Comprehensive API Resources for AI-Driven Legislative Analysis: A Report for Non-Profit Initiatives

I. Introduction and Landscape Overview

• The Imperative for Legislative Data in AI Initiatives
  Artificial Intelligence (AI) presents transformative potential for non-profit organizations engaged in legislative analysis and advocacy. By leveraging AI, these organizations can more effectively analyze voluminous legislative texts, identify emerging policy trends, predict the potential outcomes of legislative actions, and ultimately enhance their capacity to advocate for their missions. At the core of such AI-driven endeavors lies the availability of structured, comprehensive, and timely legislative data. Application Programming Interfaces (APIs) serve as the critical conduits for accessing this data, enabling automated ingestion and processing essential for training and deploying sophisticated AI models. Without robust API access to legislative information, the ability of non-profits to harness AI for deeper insights and impactful engagement in the legislative process is significantly curtailed.
• Ecosystem of Legislative Data APIs
  The ecosystem of legislative data APIs is diverse, comprising several categories of providers, each with distinct characteristics. Commercial providers, such as BillTrack50 and LegiScan, typically offer feature-rich platforms with value-added services, including sophisticated user interfaces, curated datasets, and robust API access, though these services generally come at a cost.1 Government-provided open data sources, exemplified by the U.S. Congress.gov API and various state-level data portals accessible through platforms like Data.gov, offer direct access to public legislative information, usually free of charge.3 While cost-effective, these sources may require more extensive data processing and normalization efforts on the part of the user. Thirdly, non-profit and community-driven projects, such as Digital Democracy and Open States (now part of Plural Policy under the name Plural Open), are dedicated to increasing governmental transparency by providing access to legislative data, often at no or low cost, particularly for non-commercial and research purposes.4 This varied landscape presents both opportunities and challenges for non-profits seeking to build AI tools.
  A notable characteristic of this ecosystem is the dynamic interplay between the push for open, democratized access to information and the commercial monetization of legislative data. Government initiatives and certain non-profit projects champion the principle of free public access, as seen with Data.gov and the Digital Democracy platform, which aims to provide comprehensive California legislative data freely to non-commercial users.⁵ Conversely, commercial entities are developing increasingly sophisticated platforms, often incorporating their own AI-driven analytics, and offering these as paid services. BillTrack50, for instance, provides extensive legislative tracking capabilities through a subscription model and explicitly prohibits the resale of its data by third parties, underscoring its commercial nature.¹ This dichotomy means non-profit organizations must carefully navigate their options, weighing the benefits of free, raw data—which may necessitate greater in-house technical effort for processing and analysis—against the potential efficiencies and advanced features of paid services, which come with budgetary implications and specific usage restrictions.
  Furthermore, the role of AI is evolving within this ecosystem, with some API providers moving beyond merely supplying raw data to offering AI-generated insights as part of their service. BillTrack50, for example, incorporates AI-generated bill summaries and has partnered with AI firms like Trieve to enhance features such as the discovery of similar legislative documents.¹ Similarly, Plural Policy highlights its AI-powered insights, including bill summarization and policy trend forecasting, as a core component of its offering.⁸ This trend presents a strategic choice for non-profits: they can opt to consume these pre-analyzed outputs, potentially accelerating their work but with less control over the analytical process, or they can access more fundamental data to develop their own custom AI models, offering greater control and specificity at the cost of increased development effort.
• Critical API Selection Criteria for Non-Profits
  For non-profit organizations, the selection of appropriate APIs for an AI for legislation project hinges on several critical criteria. Cost-effectiveness is paramount; this includes the availability of free tiers, substantial non-profit discounts, and transparent pricing structures. BillTrack50 offers a “Citizen tier” for free basic searching 7 and Plural Policy has a free account option, with inquiries for non-profit discounts encouraged for premium services.10 Digital Democracy stands out by offering its API free of charge to non-commercial registered users.5
  Data comprehensiveness and granularity are crucial for training effective AI models. The API must provide access to the full spectrum of relevant data, such as complete bill texts, amendment histories, detailed sponsor and co-sponsor information, voting records at various legislative stages, committee assignments and actions, and, where available, hearing transcripts or summaries. The update frequency of the data ensures that AI analyses are based on the most current information, which is vital in the fast-paced legislative environment.
  Ease of API integration and use, supported by clear documentation, well-defined schemas, and potentially client libraries, can significantly reduce development overhead. Licensing terms must be carefully reviewed to ensure they align with the non-profit’s intended use, including rights for data storage, analysis, and any potential redistribution or public-facing applications derived from the data. Finally, the long-term sustainability and reliability of the API provider are important considerations. The discontinuation of the OpenSecrets public API in April 2025, a long-standing resource for money-in-politics data, serves as a pertinent example of how access models can change, potentially impacting projects reliant on them.¹¹ Non-profits, therefore, need to assess the stability of API providers, especially those that are grant-funded or volunteer-driven, and consider contingency plans.

II. Commercial Legislative Data Platforms & APIs

• A. BillTrack50 API

• Overview: BillTrack50 positions itself as a comprehensive legislative tracking service designed for ease of use, catering to a diverse clientele including non-profits, advocacy groups, and corporate government affairs departments.⁷ A significant aspect of its offering is the increasing integration of AI-powered features to enhance bill discovery and analysis.
• Key Features & Data Points:
  The API facilitates bill and legislative tracking across all U.S. states and Congress, with a notable capability for identifying similar legislative documents in different jurisdictions [User Query]. A key AI-driven feature is the provision of AI-generated bill summaries, accessible via the aiSummary field.1 The API exposes a rich set of data points for each bill, including a unique billID, the official billName (title), actionDate for the most recent action, the total number of actions (events), detailed sponsors information (including party affiliation and an asterisk for primary sponsors), the stateBillID used by the state, the stateCode, the billProgress status (e.g., Introduced, In Committee, Passed, Dead, Signed/Adopted/Enacted), and committeeCategories.1
• Access & Technical Details:
  The BillTrack50 API is a RESTful web service, delivering responses exclusively in JSON format.1 Access necessitates an API key, which is associated with a company account and can be requested through their contact form. This key is crucial for authentication and must be included in the HTTP request header formatted as “Authorization: apikey $API_KEY”. To ensure data security, all API requests are mandated to be made over SSL (HTTPS).1
• Usage Policies & Limits:
  BillTrack50 enforces specific usage policies to maintain service performance. Rate Limits are set at 5 requests per second and a maximum of 5,000 requests within any 24-hour period. Exceeding these limits will result in an HTTP 429 error code, indicating that authentication was successful but the rate limit was surpassed.1 These limits are a critical consideration for projects requiring large-scale data ingestion, such as those typical in AI model training. Additionally, there are Results Limits; for searches yielding very large result sets, the API will only return a specified top number of results, although it does provide a count of the total number of records that matched the query.1 This truncation can impact efforts to gather comprehensive datasets. On a more permissive note, information retrieved via the API can be locally stored and cached for up to 24 hours, a provision that can help manage rate limits effectively.1 However, there are Commercial Usage Restrictions: selling the legislative data to a third party is explicitly prohibited, though commercial usage partnerships can be discussed with the company.1 This is an important consideration for non-profits contemplating the development of tools or services based on BillTrack50 data.
• Pricing & Non-Profit Considerations:
  BillTrack50 offers a tiered pricing structure. A free “Citizen” tier allows for searching and sharing of unlimited bills within the current legislative session, searching for current legislators and committees, and includes access to some AI-powered tracking tools.6 For more extensive needs, paid subscriptions (Pro, Unlimited, and Enterprise tiers) provide expanded features such as access for unlimited users, historical bill data dating back to 2011, and full API usage.9 Regarding non-profit pricing, BillTrack50’s official stance is: “We keep our pricing low for everyone; we don’t offer special pricing for certain organizations.” However, they do mention a willingness to discuss options for students working on academic projects 6, suggesting a potential, albeit limited, avenue for negotiation for specific non-profit use cases, though standard discounts are not offered.
  The “Freemium” model, while offering an entry point via the Citizen tier, presents limitations for comprehensive AI projects. The restriction to current session data and likely reduced API access compared to paid tiers would significantly hinder AI initiatives that depend on extensive historical data for training or require deeper insights into legislator staff or committee details.⁶ This effectively channels serious AI development towards the paid tiers, impacting a non-profit’s budget. The free tier appears to function more as an introductory offering or a lead-generation mechanism for their more robust, paid services.
  The imposed rate limits (5 requests per second, 5,000 per day) and the truncation of large search results are not merely technical constraints but also reflect strategic decisions by BillTrack50 to manage server load and potentially guide users with intensive data needs towards higher-tier subscriptions.¹ These limits directly influence how an AI project can ingest data, necessitating careful planning around batch processing, prioritized data fetching strategies, and strict adherence to the 24-hour local caching policy.¹ This operational approach contrasts with the “download everything at once” capability that might be available from some bulk data repositories.
• Suitability for AI Applications:
  The API is well-suited for AI applications due to its provision of structured JSON data. The inclusion of AI-generated summaries (aiSummary) can be directly utilized as a feature or serve as a baseline for comparison with custom-developed summarization models.1 The platform’s inherent capability to identify similar bills across jurisdictions is itself a direct AI application that could be leveraged.7 BillTrack50’s use of AI to provide features like these means a non-profit must evaluate whether to use these outputs directly, saving development time, or to use the platform’s data primarily as a source for training their own custom AI models. If the non-profit has highly specific analytical requirements or aims to build proprietary AI tools, they would likely treat BillTrack50 as a structured data source. However, the restriction on “selling data” 1 implies that any AI model trained extensively on this data cannot be easily commercialized if its output effectively re-packages BillTrack50’s core data. The rate and results limits, while restrictive, can be managed with careful planning and by utilizing the 24-hour caching allowance.
• B. LegiScan API

• Overview: LegiScan offers a suite of legislative data services, providing access to bill monitoring tools and legal texts in JSON format. The platform caters to a range of users through both free and paid service tiers.²
• Key Features & Data Points:
  Core functionalities include bill monitoring and the provision of legal text in JSON format [User Query]. LegiScan boasts access to a national legislative database covering all 50 U.S. states as well as the U.S. Congress.12 The data accessible through the API encompasses detailed bill information, sponsor details, legislative history, full bill text, amendments, vote records, and related datasets. This can be inferred from the general capabilities expected of a comprehensive legislative tracking service and is supported by examples found in the documentation for pylegiscan, a Python wrapper for the API.13 LegiScan also provides specific datasets for individual states, such as California 14, New York 15, and Florida 16, which are available in both JSON and CSV formats and are typically updated on a weekly basis.
• Access & Technical Details:
  The LegiScan API is designed as a back-end JSON data service.2 To utilize the API, users must register on the LegiScan website and obtain an API key.13 For developers working in Python, an interface named pylegiscan is available to interact with LegiScan’s “pull API.” It is important to note, however, that this library appears to be quite old, with its last commit dating back over 11 years 13, which may impact its current utility or compatibility. Comprehensive technical specifications, including endpoint details and data structures, are contained within the official LegiScan API User Manual, which is available for download as LegiScan_API_User_Manual.pdf.17 The information within this manual is critical for a full understanding of the API’s capabilities.2
• Usage Policies & Tiers:
  LegiScan structures its services across several tiers:

• OneVote (Free Tier): This service is intended for individuals monitoring legislation in their home state and the U.S. Congress. It is limited to tracking a maximum of 50 bills, provides weekly email alerts, offers reports only in HTML format, and is restricted to a single user account.¹²
• OneVote+ (Paid Tier): Designed for users needing to track legislation across multiple states, this tier allows for unlimited monitored bills, provides daily alerts (Monday through Friday), and enables report generation in PDF format. The cost varies based on the extent of state coverage, for example, a single state subscription is $25 per year, while national coverage is $1,000 per year.¹²
• GAITS Pro (Paid Tier): This is LegiScan’s professional-grade service, offering features such as custom client/issue tracking, advanced reporting options (including Word and Excel formats), daily alerts that include weekends, unlimited keyword searches, and support for multiple users. Pricing for GAITS Pro also varies, with single-state access starting at $100 per year, national coverage at $3,000 per year, and each additional user license costing $100 per year.¹²

This tiered access model provides a clear scalability path. However, it also means that many advanced features crucial for robust AI applications—such as unlimited bill tracking, higher frequency alerts, and broader data access capabilities—are gated behind the more expensive paid tiers. A non-profit might initiate its exploration with the OneVote free tier, but as an AI project’s data requirements grow (e.g., to include multiple states, historical data analysis, or the need for more frequent updates for real-time analytical capabilities), the associated costs will inevitably escalate. This necessitates careful strategic planning regarding which specific data elements are essential for the project and how to manage these costs effectively. The availability of bulk state datasets ¹⁴ could present a cost-effective alternative for historical data analysis if the volume of API calls becomes a significant concern.

• Pricing & Non-Profit Considerations:
  The “OneVote” free tier offers a valuable starting point for limited, individual use cases. However, its inherent limitations, particularly the cap on monitored bills and the frequency of alerts, render it likely insufficient for a demanding AI project. While the paid tiers are relatively affordably priced for single-state coverage, the cost for national coverage can become a significant expenditure for a non-profit organization. The provided documentation snippets do not mention any specific non-profit discount programs. Clarity on this would likely be found in the LegiScan API User Manual 17 or through direct inquiry with LegiScan.
  The pylegiscan library’s reliance on a “pull API” ¹³ means that the client application is responsible for periodically querying the LegiScan service to fetch updates. This contrasts with push APIs or webhook-based systems that actively notify clients of changes. For a non-profit, this implies the need to implement infrastructure for scheduling regular data polling jobs, and these polling activities will contribute to any applicable API rate limits. The frequency of these pulls will directly determine the “real-time” nature of the data within the AI system. The alerting system available in the paid tiers ¹² might help mitigate this by signaling when it is opportune to pull fresh data.
• Suitability for AI Applications:
  The LegiScan API is highly suitable for AI applications. It provides bill text in JSON format, which is an excellent input for Natural Language Processing (NLP) tasks [User Query]. The availability of structured data on bills, votes, sponsors, and other legislative elements is valuable for training various types of AI models, including predictive analytics and classification systems.13 The provision of bulk datasets for individual states 14 could be particularly beneficial for training AI models, as it allows for the acquisition of large volumes of historical data without repeatedly hitting API rate limits, assuming these datasets are comprehensive and regularly updated. While the pylegiscan library 13 indicates a developer-friendly approach from LegiScan, its age is a concern; it may require updates or serve primarily as a conceptual guide for interacting with the API. The repeated emphasis across various documents on the LegiScan API User Manual 2 strongly suggests that this manual is the definitive source for detailed technical specifications, precise data structures, and the exact terms of use. For any AI project, a thorough understanding of these details, which would be contained in the manual, is vital for designing robust data ingestion and processing pipelines.
• C. Plural Policy API

• Overview: Plural Policy presents itself as a contemporary legislative tracking tool that places a strong emphasis on AI-powered insights and a superior user experience. It directly positions itself as an advanced alternative to other platforms like BillTrack50.⁸ The platform is designed to serve a variety of users, explicitly including non-profit organizations and government relations teams.¹⁸
• Key Features & Data Points:
  Plural offers extensive legislative data covering federal legislation, all 50 U.S. states, Puerto Rico, and the District of Columbia.8 A key differentiator highlighted by Plural is its provision of real-time alerts regarding legislative changes, contrasting with the daily alert frequency of some competitors.8 The platform heavily promotes its AI-Powered Insights, which include tools for automated bill summarization, analysis of policy trends, prediction of legislative outcomes, a “Global Bill Search” feature for identifying similar bills across different jurisdictions, and a “Momentum Indicator” designed to flag bills that are likely to proceed to a vote.8 Access to more granular data such as legislative staff information, hearing details, and committee data is typically available in their premium service tiers.10
• Access & Technical Details:
  Plural provides a “powerful API” targeted at data teams, which complements its offering of bulk data downloads.18 The primary API is described as RESTful, utilizing standard HTTP methods such as GET, POST, PUT, PATCH, and DELETE.20 Authentication generally requires an API key, which is transmitted in request headers alongside other parameters like Content-Type, Authorization, Request-Timestamp, and Request-ID.20 It is worth noting that some of the detailed API technical specifications found 20 appear to describe a generic REST API structure that might be used across different Plural products (e.g., a payment gateway is referenced); however, the core offering of a legislative data API is consistently affirmed in other materials focusing on their policy tools.18 There is also mention of GraphQL in relation to Pluralsight’s API migration strategy 21, and some initial research suggested Plural Policy might offer a GraphQL API for legislative data 18, but this is not definitively confirmed for their main legislative data API in the primary documentation snippets, which lean towards a REST architecture. The “Plural Open (formerly Open States)” initiative, now part of the Plural ecosystem, also utilizes API keys for access, managed via open.pluralpolicy.com.10
• Pricing & Non-Profit Considerations:
  Plural Policy employs a tiered pricing model:

• Free Account: Allows users to search, view, and track up to a maximum of 5 bills. It also includes the ability to search and view legislator profile pages.¹⁰
• Unlimited Tracking ($59/month): This tier provides unlimited bill tracking and is designed for a single user seat. However, Plural’s AI features and collaborative tools are not included in this package.¹⁰
• Premium (Starting at $799/month): This comprehensive tier includes access to all AI-powered insights tools, legislative staff data, hearing and committee information, data export capabilities, unlimited user seats, and team collaboration features, along with premium support and policy guidance.¹⁰

Regarding non-profits, Plural Policy’s FAQ section includes the question, “Do you offer a nonprofit discount?”.¹⁰ While a direct answer isn’t provided in the snippet, users are directed to email support@pluralpolicy.com to request a free trial of the Premium plan. This suggests that options for non-profit discounts or special arrangements might be available and are likely discussed on a case-by-case basis. The free tier is quite limited (tracking only 5 bills), which would quickly push users requiring more substantial capabilities towards paid options. The significant price increase to the Premium tier, which unlocks the full suite of AI features and unlimited user access, indicates a primary focus on organizational clients.¹⁰ The explicit mention of a non-profit discount inquiry and the availability of a premium trial are positive indicators.¹⁰ Non-profit organizations should actively pursue these avenues, as the standard pricing for the full AI feature set is substantial. The integration of “Plural Open (formerly Open States)” into their offerings ¹⁰ could also potentially provide a lower-cost pathway to accessing raw legislative data, separate from the premium AI tools.

• Suitability for AI Applications:
  Plural Policy’s offerings are highly suitable for AI applications, largely due to their explicit focus on AI-powered tools and insights.8 Non-profit organizations can choose to leverage these pre-built AI features directly or use the underlying comprehensive data for developing their own custom models. The provision of real-time alerts is particularly beneficial for AI systems designed for dynamic monitoring and rapid response to legislative developments. The broad geographic coverage (federal, all 50 states, PR, D.C.) is a significant advantage for national-scope AI projects.8 While the Premium tier offers the richest dataset and the full array of AI features, its cost necessitates that non-profits explore potential discounts.
  Plural’s strategic emphasis on AI as a core value proposition, rather than an add-on feature, suggests that their data is likely structured and curated with AI applications as a primary consideration.⁸ This could translate to a lower barrier to entry for non-profits wishing to leverage sophisticated AI insights, especially if they opt to use Plural’s existing tools. However, it also implies that the “value-add” is significantly tied to their proprietary AI analytics. Accessing the truly raw, unprocessed data for custom AI model development might be a different experience than utilizing their AI-filtered outputs, and terms for such access would need clarification. The platform’s emphasis on “real-time alerts” ⁸ positions it as a strong contender for AI projects that require up-to-the-minute information for monitoring fast-moving legislative activities. This capability implies a more dynamic data infrastructure on Plural’s part. However, the precise definition of “real-time” (which can range from seconds to minutes) and the actual latency and reliability of these alerts would need empirical testing. This also has implications for a non-profit’s own infrastructure if they intend to consume and act upon these alerts in a truly real-time fashion.

III. Open Access and Government Legislative Data APIs

• A. Congress.gov API (via Data.gov)

• Overview: The Congress.gov API serves as the official, authoritative source for U.S. federal legislative information. It provides public access to machine-readable data derived from the extensive collections available on the Congress.gov website, facilitated through the Data.gov platform.³
• Key Features & Data Points:
  The API enables users to view, retrieve, and reuse a wide array of legislative data.3 Specific data available includes comprehensive bill status information, detailed bill summaries, and potentially other critical legislative collections such as committee reports, legislator profiles, and voting records. The full scope of “collections available” would be detailed in the API’s documentation.3 Complementing the API, Bill Status Bulk Data is also accessible from the Government Publishing Office’s (GPO) bulk data repository and the govinfo API. This bulk data resource includes all data from the existing Bill Summaries dataset and references the Congressional Bills dataset.23
• Access & Technical Details:
  The primary API endpoint is structured as https://api.data.gov/congress/v3?api_key=.3 Access to the API requires an API key, which can be obtained by registering on the Data.gov signup page.3 The API returns responses in either XML or JSON formats, providing flexibility for developers.3 The current operational version of the API is version 3 (v3).3 The API is maintained by Congress.gov staff, and any issues or feedback can be reported through a dedicated GitHub repository (api.congress.gov GitHub repository). This repository also serves as a crucial resource for user guides, detailed documentation on available API endpoints, and sample client code.3
• Usage Policies & Limits:
  Information regarding specific rate limits for the Congress.gov API is not detailed in the provided snippets. Such limits, which are critical for planning data ingestion strategies for AI projects, would be outlined in the official API documentation available on the GitHub repository.3
• Pricing & Non-Profit Considerations:
  Access to the Congress.gov API and its data is free of charge. This aligns with the broader mission of Data.gov to promote open government and provide public access to federal data, which is highly advantageous for non-profit organizations operating with budget constraints.
• Suitability for AI Applications:
  The Congress.gov API is exceptionally well-suited for AI applications focused on federal legislation. It provides an authoritative source for bill texts, legislative status updates, summaries, and potentially votes and sponsor information – all crucial inputs for AI models. The availability of data in standard JSON and XML formats facilitates straightforward data ingestion and processing.3 Furthermore, the provision of Bill Status bulk data 23 is a significant asset for AI projects, enabling the training of models on large historical datasets without the need to make repetitive API calls, thereby conserving resources and avoiding potential rate limit issues. The richness and structured nature of the data, such as full bill texts and detailed records of legislative actions, make it ideal for Natural Language Processing (NLP), predictive modeling, and other advanced AI analyses.
  As the official source, the Congress.gov API offers a high degree of authority and comprehensiveness for federal legislative data, making it an invaluable asset for non-profits. The free access model is a major benefit. For any AI project centered on U.S. federal legislation, this API should be considered a primary data source. Its direct connection to the legislative process helps ensure data accuracy and timeliness, although the precise update frequency of the API relative to the main Congress.gov website would need to be verified from the full documentation. The availability of bulk data downloads ²³ further enhances its utility for AI, reducing the reliance on extensive API polling for historical data.
  While the foundational aspects of the API are clear, critical operational details such as specific rate limits and the precise data points contained within the various “collections” are deferred to the extensive documentation hosted on GitHub.³ It is imperative for non-profits to thoroughly consult this external documentation. The success of leveraging this API for an AI project directly hinges on a clear understanding of these operational parameters. For instance, undefined or restrictive rate limits could become a significant bottleneck if not anticipated and managed appropriately. The exact structure and content of the “collections” will dictate how data needs to be queried, combined, and processed for AI modeling. The platform’s dual-access model, offering both an API for targeted, real-time queries and bulk data downloads for large-scale historical analysis ²³, is highly beneficial. AI projects can strategically use bulk data for initial model training and comprehensive historical reviews, then employ the API for ongoing updates or to retrieve specific information as needed. This approach optimizes both data acquisition efficiency and API resource utilization. Understanding the alignment and potential differences (e.g., in update cycles or data granularity) between these two resources—the API and the GPO bulk data—is an important aspect of planning.
• B. Digital Democracy API (CalMatters/Cal Poly Project)

• Overview: Digital Democracy, a collaborative project prominently involving CalMatters and California Polytechnic State University (Cal Poly), was launched in 2024. It is an AI-powered platform designed to significantly enhance transparency in California’s state government. The platform provides free public access to an extensive and searchable database of legislative information.²⁴ This initiative builds upon earlier work and funding involving entities such as the Institute for Advanced Technology and Public Policy (IATPP) at Cal Poly, the Knight Foundation, and Arnold Ventures.⁵
• Key Features & Data Points:
  The platform focuses exclusively on California state legislative information. Its database is remarkably comprehensive, including verbatim transcripts of “every word spoken in public hearings,” detailed records of campaign donations, information on all bills introduced, votes cast by legislators, extensive legislator profiles with voting histories, and financial disclosure data.5 Digital Democracy achieves this by integrating data from a multitude of sources, including MapLight, The National Institute on Money in State Politics, the Sunlight Foundation, official state legislative websites, the California Secretary of State’s databases of registered lobbyists, and various government ethics agencies.5 The platform itself utilizes AI technologies such as data mining, natural language processing, and notably, facial and voice recognition to identify and specify speakers and lobbyists participating in legislative hearing videos.5
• Access & Technical Details:
  An API is available for accessing the platform’s data, provided in JSON format, and is currently at version 0.5.5 This API allows programmatic access to the rich dataset that powers the main Digital Democracy platform.5 The website infrastructure involves a custom WordPress API integration that facilitates the flow of data from a central database, maintained by Cal Poly, to the public-facing website.24 It is not explicitly clear from the snippets whether the public API offered to users is the same as this internal WordPress API or a distinct interface designed for external developers. While detailed developer documentation or specific API endpoint listings beyond the general mentions were not available in all provided snippets, the Participedia entry for the project serves as a key informational resource.5 One attempt to access the main project website indicated it was inaccessible 25, but other sources confirm the API’s existence and core characteristics.
• Pricing & Non-Profit Considerations:
  A significant advantage for non-profit organizations is that the Digital Democracy API is offered free of charge to non-commercial registered users.5 This aligns with its mission to foster civic engagement and accountability. The project’s funding by philanthropic organizations like the Knight Foundation and Arnold Ventures underscores its public service orientation.5
• Suitability for AI Applications:
  Digital Democracy is exceptionally well-suited for AI applications, particularly those involving NLP, multimodal AI (combining text, audio, and potentially video analysis), and network analysis. The dataset’s richness is a key factor: full hearing transcripts, detailed voting records, campaign finance data, and comprehensive legislator activity logs provide a multifaceted view of the legislative process.5 The platform’s own sophisticated use of AI technologies (such as facial/voice recognition for speaker attribution and NLP for processing transcripts) suggests that the underlying data is structured in a way that is highly amenable to further AI processing and analysis.5 The provision of unique data points, such as verbatim records of “every word spoken” in public hearings, offers unparalleled opportunities for deep textual analysis and understanding legislative discourse at a granular level.
  The depth and nature of the data provided by Digital Democracy, especially the full hearing transcripts linked to identified speakers and lobbyists, represent a significant advancement for state-level legislative transparency and are of immense value for AI-driven research.⁵ While most legislative APIs primarily offer bill texts and associated metadata, Digital Democracy’s focus on the dynamics of hearings, the spoken word, and the ability to link this information to financial contributions and voting patterns allows for a much richer and more nuanced AI analysis of influence, debate, and decision-making processes within the California legislature. This platform could well serve as an influential model for similar transparency initiatives in other states.
  The “free for non-commercial registered users” clause is ideal for a non-profit’s internal AI project or research.⁵ However, if a non-profit intends to create a public-facing tool or service that utilizes this data, it would be prudent to seek clarification from Digital Democracy on whether such use cases fall within the permitted scope of “non-commercial use.” The definition of “non-commercial” can sometimes be ambiguous, particularly if the tool developed helps solicit donations for the non-profit or is part of a larger, externally funded initiative.
  The project’s history demonstrates a reliance on foundational grants and strong collaborations between academic institutions (Cal Poly) and media organizations (CalMatters), with support from major philanthropic bodies.⁵ While the platform is currently robust and innovative, its long-term sustainability and the potential for expansion to other states are likely contingent on continued funding and the success of these collaborative efforts. This represents a potential risk factor that non-profits should consider when building long-term AI strategies that depend heavily on this specific API. The successful evolution from an initial Cal Poly-centric project ⁵ to a broader CalMatters-led platform [User Query] is a positive indicator of its viability but also underscores this ongoing dependency. The mention of exploring the expansion of the Digital Democracy model to other states is a promising development for the future but remains contingent on securing resources and partnerships.²⁴
• C. Open States API (Plural Open)

• Overview: Open States has a long history as a valuable project dedicated to providing public access to U.S. state legislative information, including data on pending legislation and votes. This initiative is now part of Plural Policy and is referred to as “Plural Open”.⁴
• Key Features & Data Points:
  The platform aggregates data from all 50 U.S. states, the District of Columbia, and Puerto Rico, aiming for comprehensive national coverage at the state level.10 The Open States data schema is well-defined and includes core concepts such as Jurisdiction (representing a state or other legislative body), Session (legislative periods), Bill (encompassing all types of legislative documents like bills, resolutions, etc.), Vote (records of votes in chambers or committees), Person (information on legislators and other relevant individuals), Organization (representing legislatures, chambers, committees, and political parties), Post (a specific role or seat within an organization, e.g., a legislative district seat), and Membership (linking a Person to a Post for a defined period).26 Data concerning individuals (legislators) and committees is also maintained as editable YAML files within a dedicated people repository, indicating a community-contributable aspect to some data elements.27
• Access & Technical Details:
  Open States offers an API, with API v3 being the current version referenced.27 Access to this API requires an API key, which must be included in requests, typically as an x-api-key header.28 Management of these API keys for Plural Open (which incorporates Open States) is handled through the open.pluralpolicy.com portal.18 Postman documentation for the Open States API v3 provides examples of available endpoints, including 28:

• GET /jurisdictions: Lists all supported jurisdictions.
• GET /jurisdictions/{jurisdiction_id}: Retrieves detailed information for a single specified jurisdiction.
• GET /bills: Allows searching for bills using various parameters such as q for full-text search, sort options, jurisdiction, sponsor, and others.
• GET /bills/{openstates_bill_id}: Fetches detailed information for a specific bill identified by its Open States bill ID. The broader Open States project architecture comprises several components, including openstates-scrapers (responsible for data collection from official state websites), the people repository (for legislator and committee YAML data), openstates-core (the backend data model and scraper infrastructure), openstates.org (the public-facing website, which also powers a GraphQL API), and the api-v3 itself.²⁷

• Pricing & Non-Profit Considerations:
  Historically, Open States was known for offering a generous free tier of API access, reflecting its origins within the Sunlight Foundation and its mission to promote government transparency. With its integration into Plural Policy as “Plural Open,” the pricing and access model has evolved. Plural Policy’s main pricing page includes an FAQ item: “For Plural Open (formerly Open States) users: Are there costs associated with increasing my current level of API calls?”.10 This explicitly indicates that API call volume can incur costs. Specific details regarding free call quotas under Plural Open, or any special rates or provisions for non-profit organizations, would need to be obtained directly from Plural Policy.
• Suitability for AI Applications:
  The Open States API is highly suitable for AI projects that require broad access to state-level legislative data. Its structured and standardized data model 26 is particularly well-suited for ingestion into AI systems, significantly simplifying the process of working with data from diverse state sources. The historically open nature of its scrapers 27 and the community contributions to person data suggest a dataset that is both rich and continuously improving. The transition to stewardship by Plural Policy may bring benefits such as more robust infrastructure and dedicated support, but it also introduces potential cost implications that non-profits must carefully assess.
  The evolution of Open States from a community-focused project, initially under the Sunlight Foundation ⁴, to its current integration within the commercial entity Plural Policy ¹⁰, reflects a broader trend among open data initiatives seeking sustainable operational models, often through commercial partnerships or hybrid approaches. This transition likely brings advantages in terms of financial stability, potentially leading to more robust infrastructure, enhanced features, and professional technical support. However, for non-profit organizations that were accustomed to Open States’ traditionally very liberal free access policies, this integration with Plural’s commercial offerings necessitates a careful re-evaluation of potential costs and the terms of API access.¹⁰ The rebranding to “Plural Open” itself signals this new phase in the project’s lifecycle.
  A significant strength of Open States has always been its standardized multi-state data model.²⁶ This consistency across the varied and often idiosyncratic legislative processes and data formats of different U.S. states is a massive advantage for AI projects aiming to conduct comparative or national-scale multi-state analysis. Open States normalizes diverse state-specific information into a coherent schema with predictable concepts like Jurisdiction, Bill, Vote, etc..²⁶ This normalization dramatically reduces the data engineering effort that would otherwise be required to train AI models on legislative data sourced from multiple states, as the core data structure remains consistent.
  The mention of both api-v3 (apparently REST-based from Postman documentation ²⁸) and a separate openstates.org repository that powers a GraphQL API ²⁷ suggests that there might be multiple access points to the data or an evolving API strategy under Plural Open. While the Postman documentation provides details for REST API v3, the existence of a GraphQL API associated with the openstates.org website implies that developers might have a choice of interface, or that GraphQL is utilized for the main website’s data needs while the v3 REST API is intended for broader third-party data access. GraphQL can offer more flexible and efficient data fetching capabilities, allowing clients to request only the specific data they need, which is particularly relevant for optimizing data retrieval for AI applications. Non-profit organizations should seek clarification from Plural Policy regarding which API (the REST v3 or the GraphQL one) is recommended for their specific use case under the Plural Open umbrella and what the associated access terms and costs are for each.
• D. Free Law Project (CourtListener API)

• Overview: The Free Law Project is a non-profit organization dedicated to providing free and open access to primary legal materials. Its flagship project, CourtListener.com, serves as a vast repository for court dockets, judicial opinions, information on judges, and oral argument recordings.²⁹
• Key Features & Data Points:
  The platform offers several key datasets relevant to legal and legislative analysis:

• Opinions Database: An extensive and searchable collection of court orders and opinions from both state and federal courts across the United States.²⁹
• PACER Filings and Dockets: CourtListener hosts one of the largest free collections of federal court documents and dockets, sourced from the Public Access to Court Electronic Records (PACER) system via the RECAP project.²⁹
• Judge Data (Deep Judge Data): This includes structured biographical information, financial disclosure records, and other details for thousands of U.S. judges.²⁹
• Oral Argument Database: The project maintains the largest online collection of court audio recordings from various appellate courts.²⁹
• Citation Lookup and Verification API: Launched in April 2024, this API allows users to submit blocks of text (up to approximately 50 pages or 64,000 characters). It then extracts legal citations, identifies invalid or ambiguous ones, and attempts to match valid citations to opinions within its extensive database of case law.³¹
• Access & Technical Details:
  The Free Law Project provides programmatic access to its data through several mechanisms:

• APIs: Dedicated APIs are available for accessing opinions, filings, judge data, financial disclosures, and more.²⁹ The new Citation Lookup API endpoint is https://www.courtlistener.com/api/rest/v3/citation-lookup/, which accepts POST requests with text data.³¹
• Bulk Data and Replication: For users with substantial data needs (“power users”), the project offers bulk data files and options for database replication, providing more direct and comprehensive access to its datasets.²⁹
• Juriscraper: This is an open-source Python library developed and maintained by the Free Law Project. It is designed to scrape opinions and other data from hundreds of court websites, facilitating the collection of legal information that may not yet be in the CourtListener database.²⁹ While some initial attempts to access main API documentation pages were unsuccessful ³³, other project pages and announcements provide significant details about API capabilities and access methods, particularly for the newer Citation Lookup API.³¹

• Pricing & Non-Profit Considerations:
  As a non-profit entity itself, the Free Law Project is committed to providing free access to legal information. Its tools, APIs, and datasets are generally available free of charge for public, research, and non-commercial use. For very high-volume API usage or for setting up database replication, it would be advisable to consult their terms or contact them directly to understand any potential policies or best practices.
• Suitability for AI Applications:
  While the Free Law Project’s primary focus is on judicial data rather than legislative data, its resources are highly relevant and valuable for “AI for Legislation” projects that aim to analyze the broader legal context, impact, or interpretation of legislation.

• The Opinions Database provides a massive corpus of legal text, which is invaluable for training Natural Language Processing (NLP) models. These models can be used to understand legal language, perform case law analysis related to specific statutes, or identify judicial trends in the interpretation of laws.
• The Citation Lookup and Verification API ³¹ is directly applicable to AI systems that process legal or legislative documents. It can be used to validate citations found within bill texts or related analytical reports, cross-reference legislation with case law, and enhance the reliability of AI-generated legal summaries or analyses.
• Judge Data can be utilized for AI projects analyzing judicial behavior, such as identifying patterns in rulings related to certain types of laws or understanding the judicial background of judges who interpret legislation.
• The availability of bulk data makes it feasible to train large-scale AI models without the constraints of API rate limits.

The resources from the Free Law Project, particularly its extensive database of court opinions and dockets ²⁹, enable an AI project focused on legislation to extend its analytical reach into how laws are subsequently interpreted, challenged, and applied within the judicial system. Legislation does not exist in isolation; its true real-world impact is often shaped and clarified through court decisions. An AI system capable of understanding both the legislative intent (derived from bill texts, committee reports, and hearing transcripts) and the subsequent judicial interpretation (derived from case law) would possess a far more powerful and holistic analytical capability. The Free Law Project provides essential data for this latter component, facilitating the development of a more comprehensive “AI for law” system.Furthermore, the provision and active development of open-source tools like Juriscraper ²⁹ empower users not only to consume the data already aggregated by CourtListener but also to potentially contribute to the data collection effort or to scrape other relevant legal sources tailored to specific project needs. For a non-profit organization with technical expertise, Juriscraper offers a pathway to extend data gathering beyond what the API directly provides. This fosters a collaborative, open-source approach to legal data access, which aligns well with the missions of many non-profit organizations. It also means that such projects are not solely dependent on the Free Law Project’s own scraping priorities and capabilities.The recent launch of the Citation Lookup and Verification API ³¹ directly addresses a significant challenge in the field of legal AI: the accurate handling and validation of legal citations. AI models that generate or analyze legal text often struggle with citations, sometimes producing erroneous or “hallucinated” references. This new API provides an important “guardrail” ³¹ by allowing AI-generated or processed text to be checked for citation accuracy against CourtListener’s vast database. This is a critical enabling technology for building trustworthy AI-powered legal tools, including those that might analyze legislative documents that extensively reference existing case law.

IV. APIs for Lobbying and Campaign Finance Data (Contextual for Legislative AI)

• Rationale for Inclusion: A comprehensive AI analysis of the legislative process extends beyond the mere text of bills and votes. Understanding the ecosystem of influence surrounding legislation—including lobbying efforts and campaign contributions—is crucial for developing nuanced insights into legislative drivers, potential biases, and the overall political economy of lawmaking. Data from these areas can provide vital context for AI models seeking to explain or predict legislative outcomes.
• A. Federal Lobbying Data
  Two primary types of API resources provide access to federal lobbying data: official government sources and commercial data providers that aggregate and enhance this information.

1. LobbyingData.com API:

• Overview: LobbyingData.com is a commercial service specializing in U.S. federal lobbying data. It transforms raw government disclosure documents into structured, analyzable datasets, offering significant value-add for researchers and analysts.³⁵
• Data Points: The platform tracks an extensive range of information, including over 1.6 million lobbying contracts, more than 200,000 lobbying entities, and data on over 78,000 unique federally registered lobbyists and their firms. Key data points include the dollar amounts spent on lobbying, the specific government agencies contacted, and, crucially, the exact bills and general issues on which entities are lobbying. The database provides historical data spanning over 24 years, and publicly traded companies involved in lobbying are matched to their respective stock ticker symbols.³⁵ The service indicates that its proprietary algorithms extract over 44 distinct attributes from government disclosure documents.³⁸
• Access & Technical Details: Data can be delivered via a REST API or through AWS S3, with options for intraday updates for real-time needs. Supported data formats are versatile, including CSV, JSON, DTA (Stata), and PKL (Python Pickle), catering to various analytical workflows.³⁶
• Pricing & Non-Profit Considerations: LobbyingData.com operates on a subscription model for regularly updated data and also offers a one-time flat-fee model for bulk historical data purchases. Pricing is variable and depends on factors such as the number of users, the type of license required, and the chosen data delivery method (API, flat files, or GUI).³⁶ According to Datarade, a data marketplace, the cost for APIs and datasets from LobbyingData.com can range from $150 to $750 per purchase, and free data samples are available.³⁹ An academic discount is available through the Dewey platform.³⁸ Non-profit organizations should contact LobbyingData.com directly to inquire about specific pricing relevant to their use case and status.
• Suitability for AI: The rich, structured, and historically deep dataset provided by LobbyingData.com is ideal for AI applications aimed at analyzing lobbying influence on legislation, identifying key political actors, tracking spending patterns over time, and correlating lobbying activities with legislative outcomes. The historical depth is particularly valuable for longitudinal AI studies.

2. Senate Lobbying Disclosure Act (LDA) API:

• Overview: This is the official API provided by the U.S. Senate Office of Public Records, offering direct public access to federal lobbying disclosure filings made under the Lobbying Disclosure Act.⁴⁰
• Data Points: The API provides access to LD-1 (Lobbying Registration) and LD-2 (Quarterly Activity Report) filings, as well as LD-203 (Contribution Reports). These filings contain data on registered lobbyists, their clients, registrant details, financial amounts related to lobbying activities, specific issues lobbied, and government entities contacted.⁴⁰ The API also provides access to lists of constants used in filings, such as filing types, general lobbying activity issues, government entity names, countries, states, and lobbyist prefixes/suffixes.⁴⁰
• Access & Technical Details: This is a REST API (version 1), accessible over HTTPS, with data returned in JSON format. While unauthenticated (anonymous) access is possible, it is subject to stricter rate limits. Users can register for an API key to receive higher rate limits. The API key must be included in the Authorization HTTP header, prefixed by “Token” (e.g., Authorization: Token YOUR_API_KEY).⁴⁰
• Rate Limits: For clients with an API Key (Registered users), the rate limit is stated as 120 requests per minute in one part of the documentation, and 20,000 requests per hour elsewhere; clarification from the detailed documentation is advisable. For unauthenticated (Anonymous) clients, the limit is 15 requests per minute or 1,000 requests per hour.⁴⁰ The API supports pagination, with a default of 25 results per page (configurable up to 25). A notable constraint is that the Filings and Contribution Reports endpoints require at least one queryset parameter (e.g., filtering by filing year) to be passed in order to paginate results beyond the first page, a measure implemented for performance reasons.⁴⁰
• Pricing & Non-Profit Considerations: Access to the Senate LDA API is free, as it is a U.S. government service. This makes it a highly attractive option for non-profits.
• Suitability for AI: This API provides a direct, authoritative source for federal lobbying data. The JSON format and REST architecture are standard for data integration. However, the rate limits and specific pagination requirements necessitate careful planning and implementation for bulk data acquisition needed for comprehensive AI model training.
• B. State-Level Lobbying Data APIs (Illustrative Examples)

• Overview: The landscape of state-level lobbying data accessibility is considerably more fragmented than at the federal level. Disclosure requirements, the digitization of records, and the availability of APIs or bulk data downloads vary significantly from state to state. Some states offer sophisticated access, while others may only provide searchable web portals with limited data export capabilities.
• California:

• CAL-ACCESS: The California Secretary of State provides raw data for both campaign finance and lobbying activity through the CAL-ACCESS system. This data is available as downloadable, tab-delimited text files, which are updated daily. While not a direct API, these bulk data files allow users with technical expertise to create their own databases for analysis. Guides to the CAL-ACCESS data structure and fields are also provided.⁴²
• San Francisco Ethics Commission: For lobbying activity within the city of San Francisco, the Ethics Commission offers multiple access methods. These include browsing online directories, downloading reports in parsed data formats from DataSF, and querying a dedicated Lobbyist API (a SODA API).⁴³

• New York: Lobbying data for New York State, including lobbyist registrations, bi-monthly reports, and client semi-annual reports, is made available through the Open NY Data portal and is also discoverable via catalog.data.gov. The data is offered in multiple formats, including CSV, JSON, RDF, and XML. Six specific datasets related to lobbying are highlighted as being available.⁴⁴
• Texas:

• City of Austin: The City of Austin provides datasets related to lobbyist reports via its Open Data Portal. This data can be accessed using OData, allowing direct connections from tools like Excel or Tableau.⁴⁶ It is important to note that this is a local (city-level) data source, not comprehensive state-wide lobbying data.
• Texas Legislature Online (TLO): While the TLO website offers extensive legislative information, including bill tracking and alerts, a direct, comprehensive API specifically for all state-level lobbying data is not explicitly detailed in the provided information.⁴⁷ The site does mention options for file downloads, which might include some lobbying-related information.

• Pennsylvania: The Pennsylvania Department of State’s Lobbying Disclosure website allows for online filing and review of registration statements and quarterly expense reports.⁴⁸ This indicates that the data is digitized and accessible online, but an API for bulk programmatic access is not specified in the snippet.
• Suitability for AI: The suitability of state-level lobbying data for AI applications is highly dependent on the specific state and the data access mechanisms it provides. Where APIs or structured bulk data downloads exist (as in parts of California and New York), this information can be invaluable for state-focused AI projects analyzing legislative influence. However, the fragmented nature of this data landscape means that a multi-state AI analysis of lobbying would require significant data sourcing, normalization, and integration efforts.
• C. Federal Election Commission (FEC) API (OpenFEC)

• Overview: The OpenFEC API is the official public interface for accessing U.S. federal campaign finance data. It provides detailed information on candidates, political committees, and their financial filings, managed by the Federal Election Commission.⁴⁹
• Data Points: The API exposes a wealth of data, including information on candidates (names, party, office sought), committees (PACs, party committees, candidate committees), financial reports (itemized receipts, disbursements, cash on hand), election dates and reporting periods, and detailed contributor information (available through Schedule A endpoints, showing donor names, amounts, dates, etc.).⁴⁹
• Access & Technical Details: The OpenFEC API is a RESTful web service. Programmatic access requires an API key, which can be obtained by signing up at api.data.gov. For testing and exploration, a DEMO_KEY can be used. The data provided through the API is updated nightly. A Swagger (OpenAPI specification) schema is available at the /swagger endpoint on the API domain, which is useful for understanding the data models and for developers creating API wrappers.⁴⁹
• Rate Limits: By default, API keys are limited to 1,000 calls per hour. Users can request an increase to 7,200 calls per hour (which translates to 120 calls per minute) by emailing APIinfo@fec.gov. Each individual API call is limited to returning a maximum of 100 results per page, necessitating pagination for larger datasets.⁴⁹
• Pricing & Non-Profit Considerations: Access to the OpenFEC API and its data is free of charge.
• Suitability for AI: This API is essential for AI projects aiming to analyze the influence of money in U.S. federal politics and its relationship to legislative processes. The detailed financial data can be linked to legislative data (e.g., from Congress.gov) to explore potential correlations between campaign contributions and legislative behavior, voting patterns, or policy outcomes. The structured nature of the data and its regular updates make it a reliable source for such analyses.
• D. OpenSecrets API (Discontinued but Custom Data Available)

• Overview: OpenSecrets.org, run by the Center for Responsive Politics, has historically been a key non-governmental source for comprehensive data on money in U.S. politics, including campaign finance and lobbying.
• Status: It is important to note that the public APIs previously offered by OpenSecrets were discontinued as of April 15, 2025.¹¹
• Alternative Access: While the public APIs are no longer available, OpenSecrets now offers custom data solutions. They can provide datasets on political contributions, expenditures, lobbying activities, and personal financial disclosures for both federal and state levels. Access to this data is by direct inquiry to commercial@opensecrets.org.¹¹ They can provide summary data, raw data, or specialized research, with data spanning more than 30 years.⁵¹
• Pricing & Non-Profit Considerations: Since access is now through custom data requests, there will likely be associated costs. Non-profit organizations interested in obtaining data from OpenSecrets would need to contact them directly to discuss their needs and any potential non-profit pricing or arrangements.
• Suitability for AI: Historically, OpenSecrets data was extremely valuable for AI projects due to its comprehensive nature and efforts to link various money-in-politics datasets. With the discontinuation of the public API, ongoing, real-time access for AI projects is more challenging. However, their custom data solutions could still be very useful for obtaining specific historical datasets for training AI models or for conducting focused research projects, depending on the cost and terms of access.

The inclusion of lobbying and campaign finance data provides a critical contextual layer that can significantly enhance the explanatory and predictive power of AI models focused on legislation. Legislation is not drafted or passed in a political vacuum. Understanding who is attempting to influence its content (via lobbying) and who is financially supporting the legislators responsible for its passage (via campaign contributions) can reveal underlying drivers, potential biases, and the broader political dynamics at play.³⁵ An AI model that only analyzes bill text or voting patterns might miss these crucial influential factors. These contextual datasets enable the development of more sophisticated AI systems that are aware of and can account for the political economy surrounding legislative action.

However, a significant challenge, particularly for AI projects aiming for multi-state analysis of influence, is the fragmentation and varying accessibility of this type of data, especially at the state level. While federal lobbying data (via the Senate LDA API ⁴⁰) and campaign finance data (via the OpenFEC API ⁴⁹) have relatively centralized and free API access points, state-level data is markedly disparate.⁴² Each state adheres to different disclosure laws, employs different data management systems, and offers varying levels of data accessibility—ranging from well-documented APIs and bulk downloads in some states to only basic searchable web portals with no easy data export in others. This heterogeneity necessitates considerable data engineering effort for aggregation, cleaning, and normalization if a comprehensive multi-state analysis is desired. Alternatively, organizations might rely on commercial aggregators, if their budgets permit.

Similar to the legislative data landscape, the provision of contextual data (lobbying and campaign finance) involves a mix of direct government APIs, which are typically free, and commercial value-added services, which are paid. Commercial services like LobbyingData.com often provide highly cleaned, structured, and historically deep datasets that can save considerable time and effort for an AI project, but this convenience comes at a monetary cost.³⁵ Government APIs like the Senate LDA API and the OpenFEC API are free to use but may require more effort from the user in terms of data processing, handling API limitations such as rate limits and pagination requirements ⁴⁰, and potentially normalizing data from different sources. Non-profit organizations must carefully weigh their budgetary constraints against the technical effort required and the desired readiness of the data. The discontinuation of the OpenSecrets public API ¹¹ also serves as a reminder of the potential volatility in relying on single sources, even those that are well-established, and underscores the importance of assessing provider stability.

V. Comparative Analysis and Strategic Recommendations for Non-Profits

To assist non-profit organizations in navigating the complex landscape of legislative and related data APIs for their AI projects, this section provides a comparative analysis and strategic recommendations. The following tables offer a structured overview, followed by a detailed evaluation framework and tailored advice.

• Table 1: Overview of Key Legislative and Contextual Data APIs

This table provides a high-level summary. Detailed features, specific costs, and terms of use should be verified directly with the API providers.

• Table 2: Detailed Comparison of Primary Legislative Tracking APIs

This table provides a comparative overview. Specific features and limitations should be confirmed with the latest documentation from each provider.

• Table 3: Key Contextual Data APIs (Lobbying, Campaign Finance, Judicial)

This table highlights key examples. State-level lobbying data availability and access methods vary widely.

• Evaluation Criteria for API Selection (Expanded):
  When selecting APIs for an AI for legislation project, non-profits should consider the following factors in depth:

• Data Relevance & Comprehensiveness: The primary consideration is whether an API provides the specific data fields essential for the intended AI analysis. This could range from full bill texts for NLP models, detailed co-sponsorship networks for influence analysis, precise voting records for predictive modeling, committee actions for tracking legislative progress, verbatim hearing transcripts for discourse analysis, or specific lobbyist activities for contextual understanding. The historical depth of the data is also critical, as many AI models require substantial historical datasets for robust training.
• Data Quality & Accuracy: The reliability of AI outputs is directly dependent on the quality of the input data. APIs sourcing data from authoritative bodies, such as Congress.gov for federal legislative information ³, generally offer higher accuracy. It’s important to understand how data is collected, validated, and what the potential for errors or omissions might be.
• Data Structure & Format: Data should be provided in formats that are easily parsable and conducive to AI processing, such as JSON, XML, or well-structured CSV files. A well-documented and consistent data schema significantly simplifies integration. For example, BillTrack50 provides JSON responses ¹, and Congress.gov offers both JSON and XML.³
• Update Frequency & Timeliness: The legislative landscape can change rapidly. The timeliness of data updates—whether real-time, daily, or weekly—is crucial for AI applications that require current information. Plural Policy, for instance, emphasizes its real-time alert capabilities ⁸, while LegiScan’s state datasets are typically updated weekly.¹⁴
• Access & Integration: The ease with which an API can be integrated into existing workflows is a practical concern. This includes the process for obtaining API keys, the clarity of authentication mechanisms, the quality and comprehensiveness of API documentation, and the availability of client libraries (e.g., the pylegiscan library for LegiScan, though its age is a factor ¹³) or SDKs.
• Cost & Licensing for Non-Profits: Budgetary constraints are often a primary concern for non-profits. A thorough assessment of subscription costs, the limitations of any free tiers, and the availability of specific non-profit discounts is essential. Licensing terms must also be carefully scrutinized for any restrictions on data usage, storage, derivation of new datasets, or redistribution of analyses, particularly if the non-profit plans to develop public-facing tools. Examples include BillTrack50’s prohibition on reselling data ¹ and Digital Democracy’s “non-commercial use” stipulation for its free API.⁵
• Rate Limits & Scalability: API call limits (e.g., requests per second/day) can significantly impact the feasibility of large-scale data ingestion for AI model training or continuous monitoring. BillTrack50, for example, has a limit of 5,000 requests per day ¹, while the default for the OpenFEC API is 1,000 calls per hour.⁴⁹ The API’s capacity must align with the project’s data volume requirements.
• AI-Specific Features/Suitability: Some APIs offer features specifically designed for or highly conducive to AI applications. This could include pre-built AI insights like automated summaries (BillTrack50 ¹), specialized search functions (Plural’s similar bill search ⁸), or unique datasets like the verbatim hearing transcripts from Digital Democracy, which are ideal for advanced NLP [User Query].
• Provider Reliability & Support: The long-term stability and maintenance commitment of the API provider are important, especially for projects intended to operate over extended periods. The level and responsiveness of technical support, as well as community forums or issue tracking systems (like the GitHub repository for the Congress.gov API ³), also contribute to the overall viability of an API.
• Recommendations Tailored to AI Project Needs:
  The optimal choice of APIs will depend heavily on the specific goals and scope of the non-profit’s AI project.

• For Broad National Legislative Tracking (Federal + All States): A multi-API strategy is typically necessary. The Congress.gov API is the definitive source for U.S. federal legislation. For comprehensive state-level coverage, Plural Policy API (incorporating Plural Open/Open States) offers wide reach, though its cost structure for high-volume use by non-profits needs careful assessment. LegiScan API is another strong contender for state data, potentially offering more granular pricing options depending on the number of states required. BillTrack50 API also provides national coverage and should be evaluated based on its feature set and pricing relative to the project’s needs.
• For Deep Dive into a Specific State (e.g., California): The Digital Democracy API is unparalleled for in-depth analysis of the California legislature, given its rich dataset including hearing transcripts, campaign finance links, and voting records, all provided free of charge for non-commercial use.⁵ This can be supplemented with data from LegiScan API or Plural Policy API for other structured legislative data points if Digital Democracy does not cover all specific needs. State-specific raw data downloads like CAL-ACCESS ⁴² can also be valuable for campaign finance and lobbying context.
• For Projects Focused on Federal Legislative Text Analysis & Prediction: The Congress.gov API, along with its associated bulk data downloads ²³, should be the primary resource. This provides authoritative access to bill texts, summaries, and status. To enrich these analyses, contextual data from the Senate LDA API (for lobbying) ⁴⁰ and the OpenFEC API (for campaign finance) ⁴⁹ would be highly valuable.
• For Analyzing Influence (Lobbying & Campaign Finance):

• Federal Level: The Senate LDA API ⁴⁰ and OpenFEC API ⁴⁹ offer free, direct access to official disclosure data. For potentially more cleaned, structured, and historically deep federal lobbying data, LobbyingData.com API ³⁵ is a strong commercial option, though paid. If historical campaign finance data is needed beyond what OpenFEC easily provides, inquiries for custom datasets from OpenSecrets ¹¹ could be considered.
• State Level: This area is highly fragmented. A state-by-state assessment is required. For California, CAL-ACCESS ⁴² (raw data) and the San Francisco Ethics Commission API ⁴³ (for SF-specific data) are key. For New York, Open NY Data provides various lobbying datasets.⁴⁴ Other states may have their own portals or require manual data collection.
• For Projects Needing Legal Context (Court Cases & Judicial Interpretations): The Free Law Project API (CourtListener) is essential.²⁹ Its extensive opinions database provides a vast corpus for analyzing case law related to legislation, and its new citation verification API ³¹ can help link legislative analysis to judicial precedents.
• Strategies for Integrating Multiple APIs:
  When an AI project requires data from multiple API sources, a robust integration strategy is crucial:

• Modular Data Ingestion Pipeline: Develop a flexible pipeline where connectors for each API can be independently managed, updated, and scaled.
• Unified Data Model: Create a common internal data schema or data model to normalize and map data elements from different APIs. This simplifies downstream processing and AI model training by providing a consistent data structure.
• Rate Limit Management and Error Handling: Implement sophisticated mechanisms to respect each API’s rate limits, including request throttling, queuing, and backoff strategies. Robust error handling is needed to manage API downtime or unexpected responses.
• Data Caching: Implement intelligent caching strategies to store frequently accessed or slowly changing data locally, reducing redundant API calls and improving performance. This must be done in compliance with each API’s terms of service (e.g., BillTrack50 allows 24-hour caching ¹).
• Highlighting Free/Low-Cost Options:
  Non-profits should prioritize exploring free and low-cost options. These include the Congress.gov API, Digital Democracy API (for California, non-commercial use), Free Law Project API, Senate LDA API, and OpenFEC API. Additionally, the free tiers offered by commercial providers like BillTrack50, LegiScan, and Plural Policy can be valuable for initial exploration, pilot projects, or very limited use cases. However, it is critical to clearly understand the limitations of these free tiers (e.g., data volume, feature restrictions, API access constraints) as they may not be sufficient for large-scale or production-level AI applications.

The diverse nature of AI for legislation projects—spanning textual analysis, outcome prediction, influence mapping, and varying geographic focus (state vs. federal)—means that a single API will rarely suffice. Non-profits will likely need to strategically combine multiple APIs to meet their comprehensive data requirements. For instance, a project analyzing federal bill progression and its influencing factors might use the Congress.gov API for legislative data, the Senate LDA API for lobbying context, and the OpenFEC API for campaign finance insights. A study focused on understanding policy debates and influence in California would heavily leverage the Digital Democracy API and potentially CAL-ACCESS for deeper financial data. This necessity for a multi-API approach underscores the importance of a flexible and well-planned data strategy rather than a search for a single, all-encompassing “perfect” API.

In making these choices, non-profits must constantly navigate a trade-off involving cost, data comprehensiveness, and operational convenience. Free government APIs like Congress.gov offer extensive data (within their specific domain) at no monetary cost but may demand more technical effort for data processing, normalization, and integration. Commercial APIs such as BillTrack50 or Plural Policy often provide greater convenience through pre-processed data, user-friendly interfaces, and value-added features (like AI-generated summaries or predictive analytics), but these benefits come with subscription fees. The “best” API or combination of APIs is therefore highly context-dependent, hinging on the non-profit’s specific project goals, technical capabilities, and budgetary realities.

It is also important to recognize the potential “hidden costs” associated with “free” API tiers or services. While many APIs are offered free of charge for non-commercial use or have limited free tiers, non-profits may still incur costs in terms of developer time for API integration, data cleaning and transformation, managing API limitations (such as restrictive rate limits or constrained data access), and the potential future need to upgrade to paid tiers if the project’s requirements outgrow the free offerings. For example, the Digital Democracy API is free for non-commercial use but is specific to California; using its model for another state is not directly possible without a similar initiative there.⁵ The free tier of BillTrack50 is limited in scope and features.⁶ A thorough Total Cost of Ownership (TCO) analysis should, therefore, consider not just direct subscription fees but also these indirect resource investments and potential future upgrade paths.

VI. Future Outlook and Emerging Resources

• Trends in Legislative Data & APIs:
  The landscape of legislative data and APIs is continually evolving, shaped by technological advancements and a growing demand for transparency and sophisticated analytics. Several key trends are emerging:

• Increased Use of AI by Data Providers: A prominent trend is the integration of AI directly into legislative data platforms. More providers are likely to offer AI-driven summaries, predictive analytics, and advanced analytical tools, following the lead of services like BillTrack50 ⁷ and Plural Policy ⁸, which already market such capabilities. This can lower the barrier to entry for some AI applications but may also create “black box” scenarios if the underlying AI methodologies are not transparent.
• Push for Greater Transparency: Initiatives like Digital Democracy [User Query], with its focus on deep transparency in California, may serve as a model and inspire similar projects in other states or at different levels of government. The demand for open access to granular legislative data, including hearing transcripts and influence data, is likely to grow.
• Standardization Efforts: There may be a gradual movement towards greater adoption of common data schemas (such as the model used by Open States ²⁶) or API standards like the OpenAPI Specification.⁵² Such standardization would significantly improve interoperability between different data sources and simplify the integration process for developers.
• GraphQL Adoption: While RESTful APIs currently dominate the landscape, GraphQL ²¹ may see increased adoption due to its ability to provide more flexible and efficient data querying. This allows clients to request only the specific data they need, which can be highly beneficial for complex AI data requirements. The GraphQL API associated with Plural’s openstates.org ²⁷ is an early indicator of this potential shift.
• Impact of New AI Technologies:
  Rapid advancements in AI, particularly in areas like Large Language Models (LLMs), will profoundly impact the legislative data domain:

• LLMs will drive a significant increase in demand for high-quality, large-volume textual legislative data, including full bill texts, amendments, committee reports, and verbatim hearing transcripts, as these are essential for training and fine-tuning such models.
• AI will enable far more sophisticated analyses of legislative language, including nuanced sentiment analysis, detection of subtle influence patterns, advanced topic modeling, and more accurate predictions of policy impacts.
• Ethical considerations surrounding the use of AI in the legislative sphere will become increasingly prominent. Issues such as bias in training data (potentially leading to skewed analytical outcomes), the interpretability and explainability of AI models (crucial for accountability), and the responsible deployment of AI-driven decision support tools will require careful attention and robust frameworks.
• Evolving Projects & Data Sources to Monitor:
  Non-profits should keep abreast of developments in several areas:

• Expansion of Digital Democracy: The potential expansion of the Digital Democracy model to other U.S. states is a key development to watch, as it could significantly increase access to rich, state-level legislative data.²⁴
• Plural Open (Open States) Developments: The continued evolution of the Open States project under the stewardship of Plural Policy will be important, particularly regarding its API offerings, data coverage, and access terms for non-profits.
• New Government Open Data Portals: The emergence of new state or local government open data portals that offer APIs for legislative, lobbying, or campaign finance data should be monitored.
• Academic and Research Initiatives: Universities and research institutions often undertake projects that result in novel legislative datasets or analytical tools. These can sometimes transition into publicly available resources.

A notable dynamic is the emerging AI feedback loop: as AI tools for legislative analysis become more sophisticated, they will, in turn, drive demand for more granular, timely, and interconnected legislative data from API providers. If an AI model demonstrates the ability to effectively predict legislative outcomes by analyzing co-sponsorship patterns, lobbying activity, and public sentiment (potentially derived from news articles or social media linked to specific bills), then there will be a greater demand for APIs that can provide this interconnected data seamlessly and efficiently. This creates a positive feedback cycle where advancing AI capabilities push data providers to innovate and expand their offerings.Concurrently, as AI is increasingly used to analyze or even predict legislative outcomes, the demand for explainable AI (XAI) will grow significantly, especially for non-profit organizations focused on advocacy, public education, and holding institutions accountable. It will not be sufficient for an AI system to merely state, for example, that a bill has a certain probability of passing. Stakeholders, including the public and policymakers, will want to understand the underlying reasons for such predictions. This necessitates that AI models are designed to be interpretable and that the underlying data sourced from APIs is transparent, well-documented, and traceable enough to support such explanations. “Black-box” AI solutions, where the decision-making process is opaque, are likely to face skepticism and limited acceptance in sensitive areas like legislative analysis.The future of legislative data access could also see either a greater integration of currently disparate datasets—such as legislative records, lobbying disclosures, campaign finance information, and judicial rulings—via advanced APIs and the development of legislative knowledge graphs, or, conversely, a reinforcement of data silos if commercial providers focus primarily on proprietary, all-in-one solutions that do not easily facilitate linking with external data sources. The ideal scenario for comprehensive AI projects is seamless access to linked data. Platforms and standards that facilitate this, for example, by promoting the use of common identifiers (like the Open Civic Data Identifiers (OCD-IDs) mentioned by Democracy Works in the context of election data ⁵⁴), or by offering robust data linking features, will become increasingly valuable. Conversely, if critical datasets remain heavily siloed and expensive or technically challenging to integrate, it could limit the scope, depth, and ultimate impact of AI-driven legislative analysis, particularly for resource-constrained non-profit organizations.

VII. Conclusions and Recommendations

The landscape of API resources for AI-driven legislative analysis is rich and varied, offering significant opportunities for non-profit organizations. However, navigating this landscape requires a strategic approach, carefully weighing factors such as data comprehensiveness, cost, access policies, and suitability for specific AI applications.

Key Conclusions:

1. No Single API Solution: The diverse needs of AI projects in the legislative domain (ranging from federal bill text analysis to state-level influence mapping or judicial impact assessment) mean that a combination of APIs is often necessary. Non-profits must adopt a flexible data acquisition strategy.
2. Trade-offs are Inherent: A fundamental trade-off exists between cost, data comprehensiveness, and convenience. Free government APIs (e.g., Congress.gov, Senate LDA API, OpenFEC API) provide authoritative data but may require more technical effort for integration and processing. Commercial APIs (e.g., BillTrack50, LegiScan, Plural Policy) offer value-added features and convenience but come at a cost.
3. Contextual Data is Crucial: For a deep understanding of the legislative process, data on lobbying (Senate LDA API, LobbyingData.com, state sources), campaign finance (OpenFEC API), and judicial interpretations (Free Law Project API) is invaluable and can significantly enhance AI models.
4. Open Access and Non-Profit Focus: Several high-quality resources are available for free or with specific considerations for non-profits (e.g., Digital Democracy for California, Free Law Project). However, even “free” resources can have indirect costs in terms of development time and potential limitations.
5. AI is Reshaping Data Provision: Data providers are increasingly incorporating AI into their own offerings (e.g., AI summaries, predictive analytics). This offers ready-made insights but also necessitates a decision by non-profits on whether to consume these or build custom AI models using raw data.
6. Sustainability and Documentation are Key: The long-term viability of API providers and the quality of their documentation are critical for project success. The discontinuation of some APIs (e.g., OpenSecrets public API) highlights the need for due diligence.

Strategic Recommendations for Non-Profits:

1. Define AI Project Scope and Data Needs Clearly: Before selecting APIs, meticulously define the specific questions the AI project aims to answer, the geographic focus (federal, specific states, national), the types of data required (bill text, votes, sponsors, hearing transcripts, lobbying data, etc.), and the necessary historical depth.
2. Prioritize Authoritative and Free/Low-Cost Sources:

• For federal legislative data, the Congress.gov API should be the primary source, supplemented by its bulk data offerings.
• For federal lobbying data, the Senate LDA API is the official free source.
• For federal campaign finance data, the OpenFEC API is the official free source.
• For judicial context, the Free Law Project API offers invaluable free resources.
• For California-specific deep dives, the Digital Democracy API is an exceptional free resource for non-commercial use.

3. Evaluate Commercial APIs Strategically:

• Assess BillTrack50, LegiScan, and Plural Policy based on their specific features (e.g., AI insights, real-time alerts, state coverage breadth) against project needs and budget.
• Actively inquire about non-profit discounts from all commercial providers, as these are sometimes available even if not publicly advertised (e.g., Plural Policy’s FAQ suggests inquiry ¹⁰).
• Utilize free tiers or trials for initial evaluation and proof-of-concept work before committing to paid subscriptions.

4. Develop a Multi-API Integration Plan:

• Assume that multiple APIs will be needed. Design a modular data ingestion architecture that can accommodate different data sources, formats, and update frequencies.
• Invest in creating a unified internal data model to normalize data from various APIs, simplifying analysis and AI model training.
• Implement robust rate limit management, error handling, and data caching strategies tailored to each API’s specific terms and constraints.

5. Thoroughly Review API Documentation and Licensing:

• Pay close attention to API documentation for details on data schemas, endpoints, authentication, rate limits, and update cycles. For government APIs, associated GitHub repositories are often key sources of documentation.
• Carefully examine licensing terms for restrictions on data use, storage, modification, and redistribution, especially if the non-profit plans to create public-facing tools or share derived insights.

6. Consider Data Quality and Timeliness: Evaluate the update frequency of each API relative to the project’s need for data freshness. For historical analysis, ensure the API provides sufficient depth.
7. Assess Long-Term Sustainability: Consider the stability and funding models of API providers, particularly for smaller or grant-funded initiatives, to mitigate risks of service disruption.
8. Stay Informed on Emerging Trends: Monitor the evolution of legislative data APIs, the adoption of new technologies like GraphQL, and the development of new transparency initiatives or AI-driven analytical tools in this space.

By adopting a well-researched and strategic approach to API selection and integration, non-profit organizations can effectively harness the power of legislative data to fuel their AI initiatives, leading to deeper insights, more impactful advocacy, and greater contributions to the public good.

Works cited

1. Web Services | BillTrack50, accessed June 8, 2025, https://www.billtrack50.com/documentation/webservices
2. LegiScan Documentation | LegiScan, accessed June 8, 2025, https://legiscan.com/gaits/documentation
3. Congress.gov API | Additional APIs and Data Services | APIs at the Library of Congress, accessed June 8, 2025, https://www.loc.gov/apis/additional-apis/congress-dot-gov-api/
4. Free Legal Research Resources – United States: Data Sources, accessed June 8, 2025, https://guides.library.harvard.edu/c.php?g=310432&p=2072006
5. The Digital Democracy Project – Participedia, accessed June 8, 2025, https://participedia.net/case/digital-democracy-california-polytechnic-university
6. Legislative Tracking System Pricing Tiers – BillTrack50, accessed June 8, 2025, https://www.billtrack50.com/info/pricing
7. Success Story: BillTrack50 | Blog – Trieve AI, accessed June 8, 2025, https://www.trieve.ai/blog/success-story-billtrack50
8. Plural vs. BillTrack50, accessed June 8, 2025, https://pluralpolicy.com/resources/plural-vs-billtrack50/
9. BillTrack50 Pricing Explained, accessed June 8, 2025, https://www.billtrack50.com/info/blog/billtrack50-pricing-explained
10. Pricing – Plural Policy, accessed June 8, 2025, https://pluralpolicy.com/pricing/
11. OpenSecrets API, accessed June 8, 2025, https://www.opensecrets.org/api
12. Legislative Tracking Features – LegiScan, accessed June 8, 2025, https://legiscan.com/features
13. poliquin/pylegiscan: Python interface for LegiScan API – GitHub, accessed June 8, 2025, https://github.com/poliquin/pylegiscan
14. California Legislative Datasets – LegiScan, accessed June 8, 2025, https://legiscan.com/CA/datasets
15. New York Legislative Datasets | LegiScan, accessed June 8, 2025, https://legiscan.com/NY/datasets
16. Florida Legislative Datasets – LegiScan, accessed June 8, 2025, https://legiscan.com/FL/datasets
17. LegiScan API User Manual, accessed June 8, 2025, https://legiscan.com/gaits/documentation/legiscan
18. Plural Policy: AI-Powered Public Policy Tools, accessed June 8, 2025, https://pluralpolicy.com/
19. Regulatory Lifecycle Management – Plural Policy, accessed June 8, 2025, https://pluralpolicy.com/regulatory-lifecycle-management/
20. API Basics – Plural Developer Documentation, accessed June 8, 2025, https://developer.pluralonline.com/reference/api-basics
21. Migrating from REST to GraphQL APIs – Pluralsight Help Center, accessed June 8, 2025, https://help.pluralsight.com/hc/en-us/articles/24420566008084-Migrating-from-REST-to-GraphQL-APIs
22. AI Bill Tracking & Intelligence Price – Plural Policy, accessed June 8, 2025, https://pluralpolicy.com/bill-tracking-intelligence-price/
23. Using Congress.gov Data Offsite | Congress.gov | Library of Congress, accessed June 8, 2025, https://www.congress.gov/help/using-data-offsite
24. CalMatters Digital Democracy Development – Fueled, accessed June 8, 2025, https://fueled.com/work/calmatters-digital-democracy/
25. accessed December 31, 1969, https://digitaldemocracy.calmatters.org/
26. Understanding the Data – Open States, accessed June 8, 2025, https://docs.openstates.org/data/
27. Getting Started – Open States, accessed June 8, 2025, https://docs.openstates.org/contributing/
28. Open States API v3 | Documentation | Postman API Network, accessed June 8, 2025, https://www.postman.com/api-evangelist/open-states/api/1942fc00-28ce-40b0-9123-8525d4e7c491/documentation/35240-f1aa5674-d4d3-460d-b323-7d6513ac3ab2
29. Home | Free Law Project | Making the legal ecosystem more equitable and competitive., accessed June 8, 2025, https://free.law/
30. CourtListener Research and Awareness Website | Free Law Project, accessed June 8, 2025, https://free.law/projects/courtlistener
31. Combat Hallucinations and Look Up Citations with our New API | Free Law Project, accessed June 8, 2025, https://free.law/2024/04/16/citation-lookup-api
32. freelawproject/juriscraper: An API to scrape American court websites for metadata. – GitHub, accessed June 8, 2025, https://github.com/freelawproject/juriscraper
33. accessed December 31, 1969, https://free.law/apis
34. accessed December 31, 1969, https://www.courtlistener.com/api/
35. Lobbying Data | Alternative Data Source & Database on United States Lobbying and Lobbying Firms, accessed June 8, 2025, https://www.lobbyingdata.com/
36. The Leading Alternative Data Provider on Lobbying – About LobbyingData.com, accessed June 8, 2025, https://www.lobbyingdata.com/about-us
37. Plans & Pricing – Downloadable Lobbying … – LobbyingData.com, accessed June 8, 2025, https://www.lobbyingdata.com/plans-and-pricing
38. Historical US Lobbying Data Now Available for Use in Academic Research, accessed June 8, 2025, https://www.accessnewswire.com/newsroom/en/business-and-professional-services/historical-us-lobbying-data-now-available-for-use-in-academic-res-791023
39. LobbyingData.com – Pricing, Reviews, Data & APIs – Datarade, accessed June 8, 2025, https://datarade.ai/data-providers/lobbying-data/profile
40. API Documentation – v1 | Lobbying Disclosure, accessed June 8, 2025, https://lda.senate.gov/api/redoc/v1/
41. Lobbying Disclosure: Home, accessed June 8, 2025, https://lda.senate.gov/
42. Raw Data for Campaign Finance and Lobbying Activity – California Secretary of State, accessed June 8, 2025, https://www.sos.ca.gov/campaign-lobbying/helpful-resources/raw-data-campaign-finance-and-lobbying-activity
43. Lobbyist Disclosure – San Francisco Ethics Commission, accessed June 8, 2025, https://sfethics.org/disclosures/lobbyist-disclosure
44. State of New York – 7 – Dataset – Catalog – Data.gov, accessed June 8, 2025, https://catalog.data.gov/dataset/?tags=lobbyist&organization=state-of-new-york
45. Public Data | New York State Commission on Ethics and Lobbying in Government, accessed June 8, 2025, https://ethics.ny.gov/public-data
46. Lobbyist – Reports | Open Data | City of Austin Texas, accessed June 8, 2025, https://data.austintexas.gov/City-Government/Lobbyist-Reports/aahu-djdd
47. Texas Legislature Online – Texas.gov, accessed June 8, 2025, https://capitol.texas.gov/
48. Lobbying Disclosure | Department of State – Commonwealth of Pennsylvania, accessed June 8, 2025, https://www.pa.gov/agencies/dos/programs/voting-and-elections/lobbying-disclosure.html
49. OpenFEC | Federal Election Commission (FEC) | Postman API Network, accessed June 8, 2025, https://www.postman.com/api-evangelist/federal-election-commission-fec/api/6613c508-7374-4615-b6cb-3fec24b950ca
50. OpenFEC | Documentation | Postman API Network, accessed June 8, 2025, https://www.postman.com/api-evangelist/federal-election-commission-fec/documentation/19lr6vr/openfec
51. OpenSecrets, accessed June 8, 2025, https://www.opensecrets.org/
52. OpenAPI Specification v3.1.1, accessed June 8, 2025, https://spec.openapis.org/oas/v3.1.1.html
53. GraphQL – Ibexa Documentation, accessed June 8, 2025, https://doc.ibexa.co/en/latest/api/graphql/graphql/
54. Democracy Works API Documentation, accessed June 8, 2025, https://developers.democracy.works/