AI Dev-Tools Census - Tygart Media

Category: AI Dev-Tools Census

Primary-source census of independent, operating AI coding-agent, MCP infrastructure, and enterprise AI developer-tooling companies.

  • XenMCP: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    XenMCP: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    AI Dev-Tools Census · Verified Record (November 20, 2026)
    Sourced Baseline: As of November 20, 2026, XenMCP operates in open beta at xenmcp.com under ESSOTEC, LLC (Kherson, Ukraine; branded 'XenMCP by Xenaxio'). It converts databases, business tools, and local stdio MCP servers into OAuth 2.1-protected remote connectors for Claude, ChatGPT, Cursor, and VS Code.
    MACHINE-READABLE TWIN & CENSUS INDEX
    XenMCP — Verified Developer Tooling Profile
    Open the byte-faithful Markdown twin for XenMCP, browse the combined AI Dev-Tools Census llms.txt feed, or return to the full Census Index. Every claim below links to its verified primary source.

    What XenMCP Builds

    XenMCP provides managed MCP hosting and an API-key-to-OAuth 2.1 bridge so AI assistants—including mobile clients like the Claude app—can connect to databases, internal APIs, and local MCP tools without exposing raw credentials or opening inbound database ports to the public internet.

    Its catalog includes specialized connectors such as the XenMCP Docker integration (xenmcp.com/catalog/docker/, updated November 15, 2026), which pairs a small MIT-licensed read-only agent running on the user's Docker host with a remote *.xenmcp.com/mcp endpoint for inspecting container state, logs, CPU, and memory.

    Quick-Reference Fact Sheet

    FieldVerified Record (As of November 20, 2026)
    CompanyXenMCP
    Official Websitehttps://xenmcp.com/
    Primary Category / RoleHosted MCP server catalog & API-key-to-OAuth 2.1 bridge
    Census ClusterMCP Infrastructure, Servers, Gateways & Registries
    Operating StatusOperating — open free beta (limited places; *.xenmcp.com/mcp)
    Independence GateIndependent private company (ESSOTEC, LLC, Kherson, Ukraine / XenMCP by Xenaxio)
    Published Pricing SummaryFree during beta (fair call limits, no credit card); planned post-beta Free and Pro tiers not yet finalized
    Key Protocols & SurfacesHosted remote MCP endpoints (*.xenmcp.com/mcp), OAuth 2.1 bridge, MIT-licensed read-only Docker agent, Claude, ChatGPT, Cursor, VS Code
    CSV Population Rank#85 (Chunk 2 (Final) of AI Dev-Tools Census)
    Verified Record DateNovember 20, 2026

    Who It Is For and How It Works

    XenMCP targets developers and small teams connecting PostgreSQL/databases, Docker hosts, or local stdio MCP servers to Claude, ChatGPT, Cursor, and VS Code with per-user OAuth access that can be revoked individually.

    Per xenmcp.com/privacy/ (updated October 2026), management runs through app.xenmcp.com while MCP endpoints are served at *.xenmcp.com. Beta participants join without a credit card under fair call limits.

    • API key to OAuth 2.1 bridge: Keeps upstream API keys and database credentials inside XenMCP while issuing revocable OAuth access to clients.
    • Local-to-remote stdio & Docker agents: Includes an MIT-licensed read-only Docker agent for inspecting containers and logs without SSH.
    • Multi-client support: Works with Claude (desktop and mobile), ChatGPT, Cursor, and VS Code.
    • Transparent beta status: Pricing page explicitly states beta is free and planned post-beta tiers remain subject to finalization.

    Published Pricing and Commercial Availability

    On xenmcp.com/pricing/ (updated November 15, 2026), XenMCP is free during its beta with fair call limits and no credit card required. The site outlines planned post-beta Free (one server, sleeps without calls) and Pro (several servers) tiers, explicitly noting that post-beta prices are not final and will be announced before beta ends.

    Corporate Independence and Funding Context

    Per xenmcp.com and xenmcp.com/privacy/ (October 2026), XenMCP ('XenMCP by Xenaxio') is operated independently by ESSOTEC, LLC (Robocha str. 52, 73000 Kherson, Ukraine; hello@xenmcp.com), with a disclaimer that XenMCP is not affiliated with the makers of the third-party systems listed in its catalog.

    Related AI Dev-Tools Census Profiles

    Verified Primary Sources

  • UGT (Use the Geek Thing): AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    UGT (Use the Geek Thing): AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    AI Dev-Tools Census · Verified Record (November 20, 2026)
    Sourced Baseline: As of November 20, 2026, UGT (Use the Geek Thing) operates independently at usegeekthing.com (redirected from usegithubthing.com) under Cameron (Kam) Low of auscaster (Myrtle Bank, South Australia). It is a free, Apache 2.0 open-source runner and hosted MCP server that turns GitHub repositories into isolated, runnable local tools.
    MACHINE-READABLE TWIN & CENSUS INDEX
    UGT (Use the Geek Thing) — Verified Developer Tooling Profile
    Open the byte-faithful Markdown twin for UGT (Use the Geek Thing), browse the combined AI Dev-Tools Census llms.txt feed, or return to the full Census Index. Every claim below links to its verified primary source.

    What UGT (Use the Geek Thing) Builds

    UGT ('Use the Geek Thing') solves the setup barrier between discovering an open-source GitHub repository and running it locally inside an AI workflow. Through usegeekthing.com, the ugt CLI, a Claude Desktop bundle, and the hosted MCP server at mcp.usegeekthing.com, AI assistants receive tested 'recipes'—verified build and execution definitions—from the catalog at api.usegeekthing.com.

    Per usegeekthing.com/privacy (updated November 11, 2026), recipes execute locally inside UGT's own Linux engine on the user's machine. User files remain local and never transit UGT servers, while any recipe from outside the reviewed catalog prompts for explicit permission before accessing folders, devices, or the internet.

    Quick-Reference Fact Sheet

    FieldVerified Record (As of November 20, 2026)
    CompanyUGT (Use the Geek Thing)
    Official Websitehttps://usegeekthing.com/
    Primary Category / RoleOpen-source local GitHub project runner, recipe catalog & hosted MCP server
    Census ClusterMCP Infrastructure, Servers, Gateways & Registries
    Operating StatusOperating — general availability (open-source CLI/desktop runner + mcp.usegeekthing.com)
    Independence GateIndependent open-source developer product (Cameron Low / auscaster, Myrtle Bank, Australia)
    Published Pricing SummaryFree and open-source under the Apache License 2.0 (per usegeekthing.com/terms)
    Key Protocols & Surfacesugt CLI/runner, Claude Desktop bundle, hosted MCP server (mcp.usegeekthing.com/mcp), recipe API (api.usegeekthing.com), Claude Code, Cursor, Gemini CLI
    CSV Population Rank#83 (Chunk 2 (Final) of AI Dev-Tools Census)
    Verified Record DateNovember 20, 2026

    Who It Is For and How It Works

    UGT targets developers and AI desktop users on macOS (Apple Silicon and Intel), Windows, and Linux who use Claude Desktop, Claude Code, Cursor, or Gemini CLI and want to run GitHub projects (such as Docling, Firecrawl, Camelot, or rembg) without manual dependency debugging.

    Users install UGT locally or load the Claude Desktop bundle from usegeekthing.com/download, select a GitHub repository or browse usegeekthing.com/recipes, and let their AI app orchestrate setup and execution with secrets kept in the host system's secret store.

    • Isolated local Linux engine: Executes GitHub project recipes on the user's own hardware without sending user files to UGT.
    • Hosted MCP & recipe catalog: Combines mcp.usegeekthing.com and api.usegeekthing.com with local ugt CLI and Claude Desktop bundles.
    • Permission-scoped security: Reviewed catalog recipes declare network/folder/device access; uncatalogued recipes require explicit user approval.
    • Verified operator & license: Discloses Cameron (Kam) Low / auscaster (Australia) and Apache License 2.0 terms on usegeekthing.com/terms.

    Published Pricing and Commercial Availability

    Per usegeekthing.com/terms (last changed November 9, 2026), UGT is free and its source code is open source under the Apache License 2.0 (excluding the UGT name and logo). No paid plans or commercial subscriptions are published.

    Corporate Independence and Funding Context

    UGT is independently made and operated by Cameron (Kam) Low of auscaster (314 Glen Osmond Road, Myrtle Bank SA 5064, Australia). Its Terms explicitly state UGT is not affiliated with, endorsed by, or sponsored by GitHub, Inc. Note: UGT cleared Chunk 1's HOLD gate in Chunk 2 after primary /terms, /privacy, /recipes, and GitHub docs verified operator identity, license, and architecture.

    Related AI Dev-Tools Census Profiles

    Verified Primary Sources

  • usefulapi.io: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    usefulapi.io: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    AI Dev-Tools Census · Verified Record (November 20, 2026)
    Sourced Baseline: As of November 20, 2026, usefulapi.io operates independently under Alexandr Maximov (IČO 88788814, Prague, Czech Republic), providing a hosted Model Context Protocol (MCP) server registry and gateway covering 1,095 integrations for Claude, ChatGPT, Cursor, and other MCP clients.
    MACHINE-READABLE TWIN & CENSUS INDEX
    usefulapi.io — Verified Developer Tooling Profile
    Open the byte-faithful Markdown twin for usefulapi.io, browse the combined AI Dev-Tools Census llms.txt feed, or return to the full Census Index. Every claim below links to its verified primary source.

    What usefulapi.io Builds

    usefulapi.io operates a hosted catalog of remote MCP servers that let AI clients connect to third-party SaaS and developer services without local server installation. On usefulapi.io, the registry lists 1,095 integrations spanning 20 categories—including 53 Developer Tools & Infrastructure integrations, 28 Observability tools, 23 Project Management tools, 79 Payments & Billing, and 156 Commerce & Memberships.

    Each listed tool exposes a dedicated remote MCP URL (for example, https://pipedrive.usefulapi.io/mcp) that users paste into Claude, ChatGPT, Cursor, or any compatible MCP client and authorize once with their own credentials.

    Quick-Reference Fact Sheet

    FieldVerified Record (As of November 20, 2026)
    Companyusefulapi.io
    Official Websitehttps://usefulapi.io/
    Primary Category / RoleHosted MCP server registry & OAuth gateway across 1,095 third-party tool integrations
    Census ClusterMCP Infrastructure, Servers, Gateways & Registries
    Operating StatusOperating — live hosted MCP endpoints (*.usefulapi.io/mcp)
    Independence GateIndependent sole-operator business (Alexandr Maximov, Prague, IČO 88788814)
    Published Pricing SummarySelf-serve hosted MCP connector endpoints on usefulapi.io; no separate public paid tier table published
    Key Protocols & SurfacesHosted remote MCP servers (*.usefulapi.io/mcp) for Claude, ChatGPT, Cursor, and any MCP client
    CSV Population Rank#82 (Chunk 2 (Final) of AI Dev-Tools Census)
    Verified Record DateNovember 20, 2026

    Who It Is For and How It Works

    usefulapi.io serves developers, operators, and AI power users who want immediate remote MCP connectivity to external APIs (such as Linear, Pipedrive, or billing/observability tools) without provisioning or maintaining self-hosted MCP bridges.

    Per usefulapi.io/privacy/ (last updated November 20, 2026), the platform acts as a pass-through gateway using user-authorized credentials to call third-party services on the user's behalf, alongside custom integration requests.

    • 1,095 hosted MCP integrations: Cataloged across 20 categories including 53 Developer Tools & Infrastructure and 28 Observability connectors.
    • Zero local setup: Each integration provides a copy-ready remote *.usefulapi.io/mcp endpoint URL.
    • Cross-client compatibility: Designed for Claude, ChatGPT, Cursor, and standard MCP clients.
    • Verified Czech operator: Terms and Privacy policies disclose Alexandr Maximov (IČO 88788814, Prague) as operator.

    Published Pricing and Commercial Availability

    On usefulapi.io (verified November 20, 2026), users can browse all 1,095 integrations and copy *.usefulapi.io/mcp endpoints directly. A dedicated /pricing route is not published (returns HTTP 404), and no public dollar tier table is listed on the primary site.

    Corporate Independence and Funding Context

    Per usefulapi.io/terms/ and usefulapi.io/privacy/ (effective July 16, 2026; updated November 20, 2026), usefulapi.io is operated independently by Alexandr Maximov (IČO 88788814, Jeseniova 1196, 130 00 Prague, Czech Republic) and is not affiliated with or endorsed by the third-party services it wraps.

    Related AI Dev-Tools Census Profiles

    Verified Primary Sources

  • Spiderbrain: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    Spiderbrain: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    AI Dev-Tools Census · Verified Record (November 20, 2026)
    Sourced Baseline: As of November 20, 2026, Spiderbrain operates at spiderbrain.ai and spiderbrain.ai/api/ as a product of Perform Digital Private Limited (India CIN U72900UP2019PTC121932). It provides a deterministic context-layer API and MCP endpoint for blast-radius analysis, code mapping, and governed enterprise AI agent memory.
    MACHINE-READABLE TWIN & CENSUS INDEX
    Spiderbrain — Verified Developer Tooling Profile
    Open the byte-faithful Markdown twin for Spiderbrain, browse the combined AI Dev-Tools Census llms.txt feed, or return to the full Census Index. Every claim below links to its verified primary source.

    What Spiderbrain Builds

    Spiderbrain builds a deterministic context-layer engine ('Brain-as-a-Service') that ingests engineering repositories and business data into a verifiable knowledge graph exposed over a single authenticated API and MCP server. On spiderbrain.ai and spiderbrain.ai/api/, the company emphasizes deterministic memory, change-impact and blast-radius analysis, and zero-exposure code mapping for coding agents.

    Every indexed brain publishes a reproducible SHA-256 verification digest (byte-identical inputs yield the same 16-hex prefix within a given engine major version) backed by an append-only, worker-signed transparency log. Supported coding-agent integrations listed on spiderbrain.ai include Codex, Continue, Cody, Antigravity, Kiro, Devin, and Aider.

    Quick-Reference Fact Sheet

    FieldVerified Record (As of November 20, 2026)
    CompanySpiderbrain
    Official Websitehttps://spiderbrain.ai/api/
    Primary Category / RoleDeterministic context-layer API & MCP endpoint for code blast-radius analysis
    Census ClusterAgent Engineering, Observability & Infrastructure
    Operating StatusOperating — beta (API keys live; dedicated single-tenant business deployments)
    Independence GateIndependent private company (Perform Digital Private Limited, India CIN U72900UP2019PTC121932)
    Published Pricing SummaryBeta API access on spiderbrain.ai/api/; dedicated single-tenant server deployments via sales on spiderbrain.ai/business/
    Key Protocols & SurfacesAuthenticated REST API, MCP endpoint, desktop app, CLI, and agent integrations (Codex, Continue, Cody, Antigravity, Kiro, Devin, Aider)
    CSV Population Rank#78 (Chunk 2 (Final) of AI Dev-Tools Census)
    Verified Record DateNovember 20, 2026

    Who It Is For and How It Works

    Spiderbrain is built for software engineering teams and enterprise AI deployments that require consistent, auditable context across repositories without local re-indexing or shared multi-tenant storage. Developers can create an API key directly on spiderbrain.ai/api/ to connect coding agents via REST or MCP.

    For organizational workloads on spiderbrain.ai/business/, Spiderbrain provisions a dedicated single-tenant server per client rather than running a shared SaaS, resolving identity across ingested sources and gating agent actions under customer-controlled approval policies.

    • Deterministic context layer: Byte-identical inputs produce identical SHA-256 digests with a worker-signed transparency log.
    • Blast-radius & code mapping: Tracks cross-file dependencies and change impact before AI coding agents modify production code.
    • Single API & MCP surface: Connects desktop apps, CLI tools, and agents including Codex, Continue, Cody, Antigravity, Kiro, Devin, and Aider.
    • Single-tenant business isolation: Dedicated server instances provisioned per organization under spiderbrain.ai/business/.

    Published Pricing and Commercial Availability

    On spiderbrain.ai/api/ and spiderbrain.ai/business/ (verified November 20, 2026), Spiderbrain operates in beta with self-serve API key creation on spiderbrain.ai/api/ and dedicated single-tenant enterprise deployments scoped via sales. Public flat-rate dollar pricing tiers are not published on spiderbrain.ai.

    Corporate Independence and Funding Context

    Per spiderbrain.ai/terms/ and spiderbrain.ai/privacy/ (last updated June 10, 2026), Spiderbrain is operated independently by Perform Digital Private Limited, a company registered in India (CIN U72900UP2019PTC121932).

    Related AI Dev-Tools Census Profiles

    Verified Primary Sources

  • Robinson AI Systems: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    Robinson AI Systems: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    AI Dev-Tools Census · Verified Record (November 20, 2026)
    Sourced Baseline: As of November 20, 2026, Robinson AI Systems operates independently as a Colorado-based startup at robinsonaisystems.com, offering free public-beta hosted MCP servers—including GitHub Unleashed (519 tools with read-only and confirmation guardrails), Vercel Unleashed, and Workspace Unleashed—with an optional Toolkit Hub and Neon Unleashed in preparation.
    MACHINE-READABLE TWIN & CENSUS INDEX
    Robinson AI Systems — Verified Developer Tooling Profile
    Open the byte-faithful Markdown twin for Robinson AI Systems, browse the combined AI Dev-Tools Census llms.txt feed, or return to the full Census Index. Every claim below links to its verified primary source.

    What Robinson AI Systems Builds

    Robinson AI Systems builds permission-scoped hosted Model Context Protocol (MCP) connectors designed to let Claude, ChatGPT, and other MCP-compatible assistants interact safely with developer and business accounts under account-holder OAuth authority. On robinsonaisystems.com/tools, three connectors are live in public beta: GitHub Unleashed, Vercel Unleashed, and Workspace Unleashed, with Neon Unleashed and Robinson's Toolkit Hub (an optional unified connection entry point) in development.

    Its flagship GitHub connector (github.robinsonaisystems.com/api/mcp) exposes a live machine-readable status manifest listing 519 tools grouped into three permission profiles—read_only (215 tools), everyday (306 tools), and full (518 tools)—backed by GitHub OAuth sign-in rather than password sharing.

    Quick-Reference Fact Sheet

    FieldVerified Record (As of November 20, 2026)
    CompanyRobinson AI Systems
    Official Websitehttps://www.robinsonaisystems.com/
    Primary Category / RoleHosted permission-scoped MCP servers (GitHub Unleashed, Vercel Unleashed, Workspace Unleashed) & optional Toolkit Hub
    Census ClusterMCP Infrastructure, Auditors & Registries
    Operating StatusOperating — public beta (no charge, no checkout; GitHub, Vercel & Workspace MCP servers live)
    Independence GateIndependent Colorado-based software startup
    Published Pricing SummaryFree public beta across live MCP tools (no paid checkout published as of November 20, 2026)
    Key Protocols & SurfacesHosted Streamable HTTP MCP (/api/mcp), OAuth 2.1 (/.well-known/oauth-authorization-server), 519 GitHub tools across read-only/everyday/full profiles
    CSV Population Rank#77 (Chunk 1 of AI Dev-Tools Census)
    Verified Record DateNovember 20, 2026

    Who It Is For and How It Works

    Robinson AI Systems serves individual developers and early adopters who want remote OAuth-authenticated MCP access to GitHub repositories, branches, pull requests, issues, and Actions without running local stdio servers or handing assistants unrestricted write permissions.

    The company is unusually explicit about its early validation stage: robinsonaisystems.com/about states directly that the startup is still building and validating its products and does not claim an established enterprise customer track record. Sensitive GitHub operations involving production, access control, notifications, or spending require preview and user confirmation, and users can switch to read-only mode or delete their connection at any time.

    • Live public-beta MCP suite: GitHub Unleashed, Vercel Unleashed, and Workspace Unleashed live; Neon Unleashed in preparation.
    • Granular safety profiles: GitHub Unleashed enforces read-only (215 tools), everyday (306), and full (518) profiles plus confirmation gates.
    • OAuth 2.1 native: Connects via provider OAuth (/.well-known/oauth-authorization-server) with per-repository scope selection and one-click data deletion.
    • Direct or hub topology: Individual hosted MCP endpoints work standalone, with Robinson's Toolkit Hub developed as an optional single connector.

    Published Pricing and Commercial Availability

    On robinsonaisystems.com/tools and github.robinsonaisystems.com (verified November 20, 2026), all currently live MCP servers (GitHub Unleashed, Vercel Unleashed, and Workspace Unleashed) are offered in public beta at no charge, with the company explicitly noting that there is no checkout or paid subscription tier yet.

    Corporate Independence and Funding Context

    Per robinsonaisystems.com/about, Robinson AI Systems is an independent, Colorado-based software startup. No external venture funding round, investor list, or acquisition is disclosed on primary pages.

    Related AI Dev-Tools Census Profiles

    Verified Primary Sources

  • PulseMCP: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    PulseMCP: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    AI Dev-Tools Census · Verified Record (November 20, 2026)
    Sourced Baseline: As of November 20, 2026, PulseMCP operates independently at pulsemcp.com as a public Model Context Protocol (MCP) ecosystem registry indexing 21,523 MCP servers and 593 MCP clients alongside MCP Apps, technical guides, and The Agentic Loop newsletter. Directory access is free, with a live notice indicating new server submissions are temporarily paused during an ingestion rework.
    MACHINE-READABLE TWIN & CENSUS INDEX
    PulseMCP — Verified Developer Tooling Profile
    Open the byte-faithful Markdown twin for PulseMCP, browse the combined AI Dev-Tools Census llms.txt feed, or return to the full Census Index. Every claim below links to its verified primary source.

    What PulseMCP Builds

    PulseMCP maintains one of the largest independent public directories dedicated exclusively to the Model Context Protocol (MCP) ecosystem. On pulsemcp.com/servers and pulsemcp.com/clients (verified November 20, 2026), the registry catalogs 21,523 MCP servers and 593 MCP clients—explicitly distinguishing end-user MCP client applications (such as OpenCode, VS Code, and Gemini CLI) from agent-building frameworks.

    Beyond raw server and client indexing, PulseMCP publishes technical implementation guides, MCP Apps listings, use-case writeups, and a bi-weekly engineering publication (The Agentic Loop, evolved from Weekly Pulse in May 2026).

    Quick-Reference Fact Sheet

    FieldVerified Record (As of November 20, 2026)
    CompanyPulseMCP
    Official Websitehttps://www.pulsemcp.com/
    Primary Category / RoleIndependent Model Context Protocol (MCP) server directory, MCP client registry & editorial publication
    Census ClusterMCP Infrastructure, Auditors & Registries
    Operating StatusOperating — live public directory (21,523 servers & 593 clients indexed; new listing submissions temporarily paused during ingestion rework)
    Independence GateIndependent operator (co-founded by Tadas Antanavicius; listed directly on pulsemcp.com)
    Published Pricing SummaryFree public directory and newsletter; commercial partnerships/sponsorships via Work With Us inquiry (no published self-serve SaaS pricing)
    Key Protocols & SurfacesMCP Server Directory (21,523 listings), MCP Client Registry (593 clients), MCP Apps catalog, use-case guides, The Agentic Loop newsletter
    CSV Population Rank#75 (Chunk 1 of AI Dev-Tools Census)
    Verified Record DateNovember 20, 2026

    Who It Is For and How It Works

    PulseMCP serves developers, security reviewers, and team leads auditing which MCP servers and MCP-compatible clients already exist for external systems (such as GitHub, Figma, Notion, Postgres, Jira, Slack, or Playwright) before writing custom integrations from scratch.

    Evaluating teams should note an explicit operational banner live across pulsemcp.com as of November 20, 2026: new server submissions and listing edits are paused while the maintainers rework how listings are ingested and managed, though all existing directory search, filtering, client listings, and posts remain live and accessible.

    • MCP Server Directory: Searchable index displaying 21,523 MCP servers across API, database, browser, and developer categories.
    • MCP Client Registry: Curated index of 593 MCP-capable client applications with explicit framework-exclusion criteria.
    • Editorial & guides: Publishes implementation posts, use cases, and The Agentic Loop bi-weekly newsletter.
    • Operational transparency: Clearly discloses temporary submission pause during ingestion architecture rework.

    Published Pricing and Commercial Availability

    Directory browsing, search, client listings, and newsletter subscriptions at pulsemcp.com are free to access. PulseMCP does not publish self-serve SaaS pricing tiers; commercial collaborations and ecosystem partnerships are handled via its 'Work With Us' inquiry workflow.

    Corporate Independence and Funding Context

    PulseMCP operates independently (listing founder Tadas Antanavicius directly on pulsemcp.com). No institutional funding round or acquisition is published on primary pages.

    Related AI Dev-Tools Census Profiles

    Verified Primary Sources

  • Kaiva: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    Kaiva: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    AI Dev-Tools Census · Verified Record (November 20, 2026)
    Sourced Baseline: As of November 20, 2026, Kaiva (a trading name of SLATEAI LIMITED, UK Company No. 16601902, London) operates independently at kaiv.ai, building Kaiva Bridge—a managed hosting platform turning OpenAPI specs, databases, and documents into production MCP servers. Published plans are Free ($0, 10 servers, 200k tool calls/month), Pro ($19/month, 1M calls), and Enterprise.
    MACHINE-READABLE TWIN & CENSUS INDEX
    Kaiva — Verified Developer Tooling Profile
    Open the byte-faithful Markdown twin for Kaiva, browse the combined AI Dev-Tools Census llms.txt feed, or return to the full Census Index. Every claim below links to its verified primary source.

    What Kaiva Builds

    Kaiva builds managed Model Context Protocol (MCP) infrastructure through its flagship product, Kaiva Bridge (kaiv.ai/bridge). Instead of writing boilerplate server wrappers and operating OAuth gateways, developers paste an OpenAPI specification URL, connect a database, or upload documents to generate a live hosted MCP endpoint compatible with Claude, ChatGPT, Cursor, VS Code, and custom MCP clients.

    Kaiva complements its hosted control plane with an MIT-licensed open-source CLI (npx @kaiva/bridge push my-api –openapi <spec-url>) and an administrative MCP endpoint (api.kaiv.ai/api/bridge/manage/mcp) allowing coding agents themselves to list and draft servers.

    Quick-Reference Fact Sheet

    FieldVerified Record (As of November 20, 2026)
    CompanyKaiva
    Official Websitehttps://kaiv.ai/
    Primary Category / RoleManaged MCP server hosting platform (Kaiva Bridge) turning OpenAPI specs, DBs & docs into hosted MCP endpoints
    Census ClusterMCP Infrastructure, Auditors & Registries
    Operating StatusOperating — general availability (EU default SOC 2 Type II infrastructure; MIT open-source CLI)
    Independence GateIndependent UK private limited company (SLATEAI LIMITED, England & Wales Co. No. 16601902, London)
    Published Pricing SummaryFree ($0 forever, 10 MCP servers, 200k tool calls/mo), Pro ($19/mo, unlimited servers, 1M calls then $0.04/1k), Enterprise (custom)
    Key Protocols & SurfacesManaged MCP servers (OAuth 2.1 & Bearer keys), OpenAPI-to-MCP CLI (npx @kaiva/bridge, MIT), custom domains, SAML/OIDC SSO, SCIM
    CSV Population Rank#70 (Chunk 1 of AI Dev-Tools Census)
    Verified Record DateNovember 20, 2026

    Who It Is For and How It Works

    Kaiva Bridge is built for backend teams, SaaS companies, and AI builders exposing internal APIs or customer-facing tools over MCP with proper authentication, write-action approvals, and audit trails. Per kaiv.ai/privacy, production infrastructure runs in the EU by default on SOC 2 Type II infrastructure, and Kaiva states it does not train foundation models on customer data.

    The permanent Free plan covers up to 10 MCP servers, 200,000 monthly tool calls, and 3 teammates without a credit card. Pro ($19/month) unlocks unlimited servers, 1,000,000 monthly calls, branded custom domains (mcp.yourcompany.com), per-tool access control, SAML/OIDC SSO, and require-approval gates on write operations.

    • Zero-boilerplate MCP deployment: Converts OpenAPI specs, databases, or documents into hosted MCP endpoints via console or MIT CLI.
    • Production controls on Pro: Custom domains, OAuth 2.1 & SAML/OIDC SSO, per-tool access control, and write-approval gates.
    • EU-default & SOC 2 Type II infrastructure: Credentials sealed at rest; 7-day (Free), 30-day (Pro), or 1-year+ (Enterprise) audit logs.
    • Verified corporate transparency: Publishes full UK legal entity (SLATEAI LIMITED, Co. No. 16601902) and GDPR/ICO disclosures.

    Published Pricing and Commercial Availability

    On kaiv.ai/bridge/pricing (verified November 20, 2026), Kaiva publishes Free ($0, no card, no expiry, 10 MCP servers, 200,000 tool calls/month, 200 document pages, 3 team members, 7-day audit retention, Bearer or OAuth 2.1), Pro ($19/month flat, unlimited MCP servers, 1,000,000 tool calls/month then $0.04/1,000 extra calls, 20,000 document pages, 10 members, 30-day retention, custom domains, SAML/OIDC SSO, write approvals), and Enterprise (custom contract with SCIM, SIEM export, dedicated server, and 1yr+ retention).

    Corporate Independence and Funding Context

    Per kaiv.ai/privacy (updated October 24, 2026), Kaiva is a trading name of SLATEAI LIMITED, an independent company registered in England and Wales (Company No. 16601902) with registered office in London. No external venture round or acquisition is disclosed on primary pages.

    Related AI Dev-Tools Census Profiles

    Verified Primary Sources

  • AgentDiet: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    AgentDiet: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    AI Dev-Tools Census · Verified Record (November 20, 2026)
    Sourced Baseline: As of November 20, 2026, AgentDiet operates independently from San Francisco (engineered by Roberto Yanguas) across agentdiet.dev and github-audit.agentdiet.dev, offering an MCP/A2A context-distillation network and automated GitHub repository dependency, SBOM, secret-leak, and license auditing for coding agents. Plans span Free Sandbox, $0.002/call x402, $29/month Pro, and $199/month Enterprise.
    MACHINE-READABLE TWIN & CENSUS INDEX
    AgentDiet — Verified Developer Tooling Profile
    Open the byte-faithful Markdown twin for AgentDiet, browse the combined AI Dev-Tools Census llms.txt feed, or return to the full Census Index. Every claim below links to its verified primary source.

    What AgentDiet Builds

    AgentDiet develops machine-to-machine (B2A) infrastructure designed to shrink context-window bloat and supply deterministic security checks to autonomous coding agents. On agentdiet.dev and github-audit.agentdiet.dev, the root endpoints serve live JSON discovery manifests linking OpenAPI 3.1 specs, Streamable HTTP MCP JSON-RPC endpoints (/mcp), A2A agent cards, and /llms.txt indexes.

    Its GitHub-Audit node—the entry indexed in the official MCP Registry—performs automated supply-chain, SBOM dependency, CI/CD workflow, secret-leak, and license-compliance audits on repositories for agents running inside Cursor, Claude Code, or LangGraph, while the core AgentDiet distiller converts noisy web pages into trimmed Markdown or BM25 excerpts.

    Quick-Reference Fact Sheet

    FieldVerified Record (As of November 20, 2026)
    CompanyAgentDiet
    Official Websitehttps://agentdiet.dev/
    Primary Category / RoleMCP/A2A context-window distiller & coding-agent GitHub supply-chain security auditor
    Census ClusterMCP Infrastructure, Auditors & Registries
    Operating StatusOperating — production v7.0.16 (MCP, A2A, OpenAPI, CLI & x402 micropayments)
    Independence GateIndependent operator (engineered by Roberto Yanguas, San Francisco, CA)
    Published Pricing SummaryFree Sandbox ($0), Pay-Per-Call x402/MPP ($0.002/call), Pro Developer ($29/mo, 20k calls), Enterprise SLA ($199/mo, 150k calls)
    Key Protocols & SurfacesStreamable HTTP MCP, Linux Foundation A2A v0.3.0, OpenAPI 3.1, HTTP 402 / x402 v2 / L402 / IETF MPP, CLI (@agentdiet/cli), /llms.txt
    CSV Population Rank#65 (Chunk 1 of AI Dev-Tools Census)
    Verified Record DateNovember 20, 2026

    Who It Is For and How It Works

    AgentDiet targets developers building autonomous coding-agent pipelines who want deterministic pre-install repository security checks or lower LLM token consumption, including keyless programmatic settlement via HTTP 402 micropayments.

    Developers can test schemas on the Free Sandbox (/v1/try/github-audit), invoke endpoints per call via x402 v2 (USDC across Base, Arbitrum, Polygon, Optimism, and Avalanche), IETF MPP, or L402 Lightning macaroons, or install the official CLI (npx @agentdiet/cli, pip install agentdiet-cli, brew install agentdiet) on standard monthly Stripe-metered tiers.

    • GitHub-Audit MCP node: Automated SBOM dependency, CI/CD workflow, secret-leak, and license compliance auditing.
    • Multi-protocol discovery: Implements Streamable HTTP MCP, A2A v0.3.0, OpenAPI 3.1, and /llms.txt natively.
    • Dual billing rails: Standard monthly API/MCP plans alongside keyless HTTP 402 micropayments (x402 v2, MPP, L402).
    • Multi-channel CLI: Distributed via npm (@agentdiet/cli), PyPI (agentdiet-cli), and Homebrew (agentdiet).

    Published Pricing and Commercial Availability

    On github-audit.agentdiet.dev/pricing (verified November 20, 2026), AgentDiet lists Free Sandbox ($0/month, deterministic schema preview at /v1/try/github-audit), Pay-Per-Call ($0.002/call via keyless x402 v2 or IETF MPP), Pro Developer ($29/month, 20,000 API & MCP tool calls/month with Stripe metered billing), and Enterprise SLA ($199/month, 150,000 calls/month, dedicated Cloud Run concurrency, custom A2A swarm workflows, 99.95% uptime SLA).

    Corporate Independence and Funding Context

    On github-audit.agentdiet.dev/about, the service is identified as part of the 31-domain AgentDiet & X402-Broker network engineered independently by Roberto Yanguas in San Francisco, CA. No external institutional funding or acquisition is disclosed.

    Related AI Dev-Tools Census Profiles

    Verified Primary Sources

  • Pydantic: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    Pydantic: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    AI Dev-Tools Census · Verified Record (November 20, 2026)
    Sourced Baseline: As of November 20, 2026, Pydantic operates independently at pydantic.dev as an end-to-end Python AI engineering stack spanning Pydantic Validation, Pydantic AI, Pydantic Evals, Logfire (OpenTelemetry observability), AI Gateway, and Agent Sandboxes. Logfire includes 10M free telemetry records/month forever, then $2 per million records on paid plans.
    MACHINE-READABLE TWIN & CENSUS INDEX
    Pydantic — Verified Developer Tooling Profile
    Open the byte-faithful Markdown twin for Pydantic, browse the combined AI Dev-Tools Census llms.txt feed, or return to the full Census Index. Every claim below links to its verified primary source.

    What Pydantic Builds

    Pydantic evolved from Python's ubiquitous runtime data-validation library (over 1 billion monthly PyPI downloads and 64K+ GitHub stars across its open-source repositories, per pydantic.dev/about) into a unified AI engineering stack. Its core products are Pydantic AI (a typed agent framework supporting 20+ model providers, durable execution via Temporal/DBOS/Prefect/Restate, and multi-agent patterns), Pydantic Evals, and Pydantic Logfire.

    Logfire is built natively on OpenTelemetry to trace agent prompts, tool calls, microservices, and host metrics on one correlated timeline, and also includes an AI Gateway (supporting BYOK with zero markup for up to 3 providers, or built-in providers with a 5% markup) and newly launched Agent Sandboxes for executing agent-authored code.

    Quick-Reference Fact Sheet

    FieldVerified Record (As of November 20, 2026)
    CompanyPydantic
    Official Websitehttps://pydantic.dev/
    Primary Category / RoleType-safe Python AI agent framework (Pydantic AI), Pydantic Evals, Logfire OTel observability & Agent Sandboxes
    Census ClusterVector Retrieval & Multi-Agent Orchestration
    Operating StatusOperating — general availability (Cloud, Dedicated & Self-hosted Enterprise + OSS)
    Independence GateIndependent private company (created by Samuel Colvin)
    Published Pricing SummaryLogfire Personal Free forever (10M telemetry records/mo, $20 value), then $2 per million records on Team/Growth; Enterprise custom
    Key Protocols & SurfacesPydantic Validation, Pydantic AI, Pydantic Evals, Logfire (OpenTelemetry-native), AI Gateway (BYOK), Agent Sandboxes, Monty, Temporal/DBOS/Prefect durable execution
    CSV Population Rank#58 (Chunk 1 of AI Dev-Tools Census)
    Verified Record DateNovember 20, 2026

    Who It Is For and How It Works

    Pydantic serves Python and TypeScript engineering teams who prefer strict static/runtime typing, OpenTelemetry standards, and composable open-source libraries over proprietary agent abstractions. Logfire instruments not only Pydantic AI, but also LangChain, the OpenAI SDK, and any OpenTelemetry-emitting service.

    Pricing uses a single record-based meter across AI observability, evaluations, APM, and infrastructure telemetry—giving solo developers 10M free records/month hard-capped at $0, with Team, Growth, Dedicated, and Self-Hosted Enterprise options.

    • Typed agent engineering: Pydantic AI brings strict schema validation, toolsets, and durable execution integrations across 20+ LLM providers.
    • OpenTelemetry-native Logfire: Correlates agent traces, evaluations, services, and Kubernetes/host telemetry in one timeline.
    • Unified record-based billing: 10M free records/month ($20 value) included on every plan, then $2/million records on Team/Growth.
    • AI Gateway & Agent Sandboxes: BYOK routing (no markup up to 3 providers) and isolated sandbox execution built into Logfire.

    Published Pricing and Commercial Availability

    On pydantic.dev/pricing (verified November 20, 2026), Pydantic Logfire lists Personal ($0 forever, hard-capped at 10M telemetry records/month across 3 projects, 30-day retention, 1 seat + 2 read-only guests, BYOK AI Gateway up to 3 providers, free Agent Sandboxes trial), Team and Growth tiers ($20/month free allowance covering 10M records, then $2 per million records; optional built-in AI Gateway providers at 5% markup), and Cloud/Dedicated/Self-hosted Enterprise plans.

    Corporate Independence and Funding Context

    Pydantic operates as an independent private company led by creator Samuel Colvin and is a member of the Open Source Pledge (per pydantic.dev/about). Primary company pages do not publish private valuation tables.

    Related AI Dev-Tools Census Profiles

    Verified Primary Sources

  • CrewAI: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    CrewAI: AI Dev-Tools Census Profile — Architecture, Pricing & Verified Record

    AI Dev-Tools Census · Verified Record (November 20, 2026)
    Sourced Baseline: As of November 20, 2026, CrewAI operates independently at crewai.com, maintaining the open-source CrewAI multi-agent Python framework (Agents, Crews, Flows) alongside its enterprise build, discovery, and runtime platform (AMP). Published pricing includes Basic (Free, 50 workflow executions/month) and custom Enterprise deployments.
    MACHINE-READABLE TWIN & CENSUS INDEX
    CrewAI — Verified Developer Tooling Profile
    Open the byte-faithful Markdown twin for CrewAI, browse the combined AI Dev-Tools Census llms.txt feed, or return to the full Census Index. Every claim below links to its verified primary source.

    What CrewAI Builds

    CrewAI builds an open-source multi-agent orchestration framework and a centrally governed enterprise runtime for coordinating teams of role-specialized AI agents. On crewai.com and docs.crewai.com, CrewAI structures developer builds around Agents (composed with tools, memory, knowledge, and Pydantic-typed structured outputs), Crews (collaborative agent groups), and Flows (event-driven deterministic orchestration with guardrails).

    For organizations scaling beyond code-only prototypes, CrewAI's commercial platform adds CrewAI Studio (a visual editor and AI copilot integrated with GitHub), CrewAI Discovery (analyzing operational tickets and workflows to rank automation candidates), and enterprise runtime governance.

    Quick-Reference Fact Sheet

    FieldVerified Record (As of November 20, 2026)
    CompanyCrewAI
    Official Websitehttps://crewai.com/
    Primary Category / RoleOpen-source multi-agent orchestration framework (Crews & Flows) & enterprise AMP build/runtime platform
    Census ClusterVector Retrieval & Multi-Agent Orchestration
    Operating StatusOperating — general availability (v1.15.44 docs; Cloud, VPC & self-hosted Enterprise)
    Independence GateIndependent private company (crewAIInc)
    Published Pricing SummaryBasic (Free, 50 workflow executions/mo + visual Studio) and Enterprise (Custom, VPC/self-hosted, SSO/RBAC/PII redaction)
    Key Protocols & SurfacesOpen-source Python SDK (uv/CLI), Agents/Crews/Flows, Pydantic structured outputs, AMP visual Studio, Discovery, /llms.txt docs
    CSV Population Rank#57 (Chunk 1 of AI Dev-Tools Census)
    Verified Record DateNovember 20, 2026

    Who It Is For and How It Works

    CrewAI targets both Python engineers authoring multi-agent systems via CLI (uv-based installation) and enterprise platform teams trying to unify business-team automation requests under central security policies. CrewAI's homepage reports usage across 65% of the Fortune 500.

    Teams can evaluate the visual Studio and GitHub integration on the free Basic plan (50 executions/month), while Enterprise unlocks SSO, RBAC, workload identity, PII redaction, private agent repositories, VPC or customer-infrastructure deployment, and optional 45-day onboarding.

    • Open-source core: Python framework (crewAIInc/crewAI) combining autonomous Crews with event-driven Flows and built-in guardrails.
    • Visual & code parity: Studio visual editor with GitHub integration alongside full CLI/SDK authoring.
    • Enterprise governance: SSO, RBAC, workload identity, PII redaction, policy enforcement, and private agent repos.
    • Flexible hosting: Deploy on CrewAI Cloud, customer VPC, or self-hosted infrastructure, backed by /llms.txt docs.

    Published Pricing and Commercial Availability

    On crewai.com/pricing (verified November 20, 2026), CrewAI publishes two commercial tiers: Basic (Free, visual Studio editor and AI copilot, GitHub integration, standard tools/triggers, 50 workflow executions/month) and Enterprise (custom pricing adding SSO, RBAC, workload identity, PII redaction, enterprise connectors, private agent repos, CrewAI Cloud / VPC / on-premises deployment, and 45-day onboarding options), alongside its free open-source framework.

    Corporate Independence and Funding Context

    CrewAI (crewAIInc) operates as an independent private company. While external funding trackers report a 2026 Series B, primary product pages at crewai.com do not publish private valuation disclosures.

    Related AI Dev-Tools Census Profiles

    Verified Primary Sources