Early access · Built for Meta's Muse ecosystem

Your REST API, ready for every AI agent.

Connector Foundry converts the API you already have into a tested, reliable agent connector. Bring an OpenAPI spec — or just a docs URL. We propose the tools, prove them in a sandbox eval, and ship only what you approve.

DEMO WALKTHROUGH — GITHUB REST API
612 operations ingested · 8 tools proposed · 12-scenario eval harness · signed v3 release candidate

Connector Foundry console dashboard showing the GitHub REST API connector pipeline: 5 of 8 tools approved, 75% eval pass rate, release blockers queue
Snapshot of the customer console running on mocked GitHub API data
01 · How it works

From your REST API to a signed agent connector, in seven gates.

Every stage is a checkpoint, not a black box. The service proposes — you review and approve. Nothing reaches an agent without your sign-off.

01

Import

BRING WHAT YOU HAVE

OpenAPI 3.x is preferred, but not required. Drop in Swagger 2.0, a Postman collection — or just a docs URL with sandbox access, and we probe the API to infer the spec. Import findings surface immediately: recursive schemas, undocumented error codes, scope gaps.

612 ops · 428 schemas 5 findings flagged at import docs-URL probing — no spec needed
02

Normalize

ONE CANONICAL MODEL

Everything compiles into a platform-neutral tool model — effects (read / write / destructive), idempotency, auth schemes, error mappings. Your connector is designed once, against no single platform's quirks.

effect-tagged operations idempotency inference no platform lock-in at design
03

Propose

TOOLS, DRAFTED FOR YOU

The tool compiler drafts agent-ready tools with plain-language intents, input schemas, parameter bindings, retry policies, and recovery mappings. Destructive operations are flagged for your decision — never smuggled into the connector.

8 tools proposed get_repository · create_issue · merge_pull_request destructive ops require ALWAYS-confirm
04

Review

YOUR DECISION QUEUE

Ambiguities land in a decision queue with evidence, impact, and a recommendation — not a config file. Should merging a PR require confirmation? What does "merge it" mean with no PR number? You decide; the connector records it.

5 open questions evidence + impact + recommendation every decision is auditable
05

Evaluate

PROOF, NOT PROMISES

An automated harness runs your connector against a frozen sandbox: tool selection, argument handling, safety gates, execution, and recovery. Failures become review questions with traces — never silent auto-patches.

12 scenarios · 75% pass selection · arguments · safety · execution · recovery full failure traces
06

Publish

SIGNED, VERSIONED, SHIPPABLE

The release manager freezes a versioned, signed bundle — OpenAPI facade, MCP server bundle, and the Muse connector adapter — ready for Meta's submission pipeline. A release checklist gates every publish: blockers resolved, eval green, approvals complete.

signed release v3 immutable bundle · sha256 digest Muse adapter contract verified
07

Operate

WATCH IT BEHAVE

Data-plane telemetry keeps watch after launch: traffic, errors, latency, prompt→tool attribution, and drift alerts when your API changes. Suggested fixes route straight back into curation — the connector improves instead of rotting.

57,028 calls / 14 days prompt attribution spec-drift detection
02 · Live demo

See it running on a real API.

These are actual snapshots of the customer console, running on mocked data from the GitHub REST API walkthrough — 612 operations in, a signed v3 connector candidate out. Click through the seven screens.

Dashboard: pipeline KPIs and release blockers for the GitHub connector
console.connectorfoundry.com/dashboard

Command center

One screen answers "can we ship?" — pipeline KPIs, the next blocking step, and every open blocker with its owner.

  • 5 of 8 tools approved, 3 awaiting your review — the pipeline never moves without you.
  • 75% eval pass rate (9 of 12 scenarios) — every failure links to its trace and fix.
  • 10 open blockers enumerated: 2 findings, 5 questions, 3 tools — "nothing ships without sign-off."
Import screen: GitHub REST API source summary and import findings
console.connectorfoundry.com/import

Bring what you have

Submit an OpenAPI spec, Swagger, Postman collection — or a docs URL. The importer inventories every operation and flags what needs your attention before a single tool is drafted.

  • GitHub REST API: OpenAPI 3.1.0, 612 operations, 428 schemas, host verified.
  • Blockers caught early: 4 recursive schemas over the depth limit, an undocumented 409 on merge.
  • Warnings with teeth: 9 POSTs lack idempotency docs, 2 ops need scopes beyond the OAuth grant.
Tools screen: curated agent tools with effect badges and approval status
console.connectorfoundry.com/tools

Curate the toolset

Every proposed tool shows its intent, effect, confirmation policy, and error recovery. You include, edit, or exclude — destructive operations are fenced, never hidden.

  • merge_pull_request is tagged DESTRUCTIVE with ALWAYS-confirm — v3 made it stricter than v2.
  • DELETE /repos/{owner}/{repo} is excluded outright: "irreversible admin action — outside agent scope."
  • Each tool carries retry policy + error mappings: 409 → don't retry, check mergeable state, escalate.
Review screen: decision queue with five open questions
console.connectorfoundry.com/review

Decide with evidence

Where the spec is ambiguous or the eval found a gap, you get a question — with evidence, impact, and a recommendation. Pick an option; the connector records your decision.

  • "What does 'merge it' mean with no PR number?" — the eval caught the agent guessing PR #482.
  • "How should the agent handle a 409 on merge?" — live probing confirmed the undocumented response.
  • Auth and scope decisions too: grant actions:write, or keep release tooling scoped down?
Evaluate screen: 12-scenario harness run with 75% pass rate and failure traces
console.connectorfoundry.com/evaluate

Proof, not promises

The harness runs your connector against a frozen sandbox and scores five dimensions. Failures ship with millisecond traces and a proposed fix — routed to your review queue, never auto-applied.

  • Run eval-2026-09-27-01: 9/12 pass — selection 2/3, arguments 2/2, safety 1/2, execution 2/3, recovery 2/2.
  • "Close without confirmation": the agent skipped the WRITES gate — fix keys confirmation to effect, not vibes.
  • "Retry storm on 500": 4 retries, duplicate drafts — fix consults the dedupe cache before every retry.
Publish screen: v3 release candidate checklist and packaging targets
console.connectorfoundry.com/publish

Ship with a signature

A release checklist gates every publish — blockers resolved, eval green, review queue empty, approvals complete, adapter contract verified. Then the bundle is frozen and signed.

  • Release v3 candidate: 8 tools, 1 breaking change, sha256 digest — immutable once signed.
  • Three targets from one review: OpenAPI facade, MCP server bundle, Muse connector adapter.
  • The checklist is honest: 5/8 approvals, eval at 75% — this candidate is blocked, and it says so.
Operate screen: traffic telemetry, prompt attribution, and drift alerts
console.connectorfoundry.com/operate

Watch it behave

After launch, the data plane reports what agents actually do with your API — which prompts become which calls, where they fail, and what your API changes break.

  • 57,028 calls in 14 days, 1.91% errors, 456ms p95 — with prompt→tool attribution per call.
  • Unmapped prompts are gap candidates: "merge the release branch into main" ×23 → propose a new tool.
  • Drift alerts: 4 new endpoints in the 2026-10-01 draft, a deprecated preview header — re-import before v4.

Snapshots show mocked demo data from the GitHub REST API walkthrough — the console ships with this demo project so you can evaluate the workflow before connecting your own API.

03 · Why Connector Foundry

Conversion is commoditized. Trust isn't.

Turning an OpenAPI spec into an MCP server is a solved problem — open-source CLIs do it, and platforms are absorbing it. We don't compete on conversion. We compete on the trust infrastructure around it: proof it works, auth your security team will sign, and an approval trail that survives procurement.

Eval harness, not a playground

No competitor offers an automated reliability certification for API owners: sandboxed scenarios across selection, arguments, safety, execution, and recovery — with traces, gates, and version-gated releases. Manual playgrounds don't count.

Spec inference from a docs URL

Every vendor we surveyed requires a clean spec upload. Point us at your docs or sandbox and we probe and infer the spec — then show you exactly what we found before anything is generated.

Propose → review → approve → publish

No formal human-approval pipeline exists anywhere else in this market. The service proposes; you dispose. Every tool, error mapping, and confirmation policy waits for explicit sign-off — nothing ships silently.

Capability Connector Foundry Spec→MCP generators
Gram · Scalar · Zuplo · OSS CLIs
Integration platforms
Composio · Arcade · Nango
Platform-native intake
Meta · Anthropic · AWS
Spec inference from docs URL / sandbox probe ✓ Yes — probe & infer, then show our work ✗ Clean spec upload required ✗ Prebuilt integrations only ✗ "Raw API + optional OpenAPI," you do the work
Automated reliability eval & certification ✓ Yes — sandboxed harness, 5 dimensions, gates ~ Manual playgrounds, no harness ✗ Evals serve agent builders, not API owners ~ Opaque gatekept testing on submission
Propose → review → approve → publish ✓ Yes — formal approval pipeline, auditable ~ Partial — PR previews, no approval gate ✗ Scoped access controls only ✗ Their review, not yours
Platform-neutral canonical core ✓ Yes — design once, adapters per platform ✗ Single artifact, usually MCP ✗ MCP-focused tool calling ✗ Their platform only
API-key → per-user OAuth auth bridge ✓ Yes — credential broker, secrets never in artifacts ~ Bundled in their hosted runtime ✓ Yes — but built for agent builders ~ Their platform's auth plumbing
Built for API owners without AI teams ✓ Yes — managed, trust-first service ~ Self-serve dashboards ✗ Agent-builder customer ~ Self-serve intake forms
Our honest take, from our own market analysis (Sept 2026): this space is crowded and consolidating — Stainless was acquired by Anthropic, Fern by Postman, and Meta already accepts "raw API + OpenAPI" in its intake. If a platform ships one-click docs→connector, standalone conversion value goes to zero. That's why we're built to survive that outcome: we're the tooling you use to produce the tested, approved artifact you submit everywhere — not a walled-garden alternative to the directories.
04 · Who it's for

The underserved middle.

Mid-market SaaS, this is your lane.

You want to be agent-ready — your customers are asking for it — but you don't have an AI team to build and babysit connectors.

Enterprise has gateways and six-figure platform deals. The low end has free tiers racing to zero. Nobody serves the middle with a managed, trust-first service priced below enterprise deals. That's us.

  • You own a REST API with real customers.CRM, billing, devtools, support, logistics — if agents should act on it, we make it safe.
  • Your security team has opinions.Per-user OAuth, vaulted secrets, redacted telemetry, auditable approvals — built for their checklist.
  • Not for you if…you're an agent builder shopping for 1,500 prebuilt integrations — that's Composio/Arcade territory, and they're good at it.
05 · Pricing

Simple tiers, buyer-friendly metering.

Priced on telemetry lookback and project count — not per tool call. Early-access pricing is set at onboarding; these are the packaging levers.

Starter

Your first connector, end to end.

Early access
custom pricing at onboarding
  • 5 API projects
  • 3-month telemetry lookback
  • Full pipeline: import → eval → publish → operate
  • Eval harness & approval workflow included
  • Muse connector adapter
Request access

Scale

Agent-ready as a platform strategy.

Early access
custom pricing at onboarding
  • Unlimited API projects
  • 12-month telemetry lookback
  • Everything in Growth
  • Multi-platform adapters as they ship
  • Dedicated support & onboarding
Talk to us

Seats, support tier, and eval depth are set at onboarding · Lookback windows beyond your plan show as locked with the required tier named

06 · FAQ

Straight answers.

No. OpenAPI 3.x is preferred, but we also accept Swagger 2.0, Postman collections, or just a docs URL with sandbox access — we probe your API and infer the spec, then show you exactly what we found before generating anything. No other vendor productizes this.
No — it feeds them. Meta's intake accepts "raw API + optional OpenAPI" and runs its own checks; Anthropic owns a generator and a marketplace. We produce the tested, approved artifact you submit everywhere — eval-gated, signed, and versioned — so you're not hostage to any single platform's roadmap.
v1 ships Muse connectors. Underneath is a platform-neutral canonical core, so OpenAI (function calling / GPT Actions / Apps SDK) and Anthropic (MCP / tool use) adapters ride the same reviewed design — no re-curation per platform.
You do. The service proposes tools, confirmation policies, and error handling; you review and approve each one. Destructive operations are flagged or excluded, writes require confirmation gates, and nothing publishes until your review queue is empty.
Through a credential broker: your API keys live in Secrets Manager + KMS and never enter release artifacts. End users connect with their own OAuth grants — per-user credentials, vaulted and refreshed — so one customer's token is never another's.
The Operate screen watches for spec drift: new endpoints, deprecations, and policy changes surface as alerts with one-click re-import and re-eval. Suggested fixes route back into tool curation — your connector improves instead of rotting.
Telemetry is redacted before logging — secrets and PII never hit the pipeline. Retention follows your tier: 3, 6, or 12 months. Windows beyond your plan appear locked in the console with the required tier named.
07 · Early access

Be agent-ready before your customers ask twice.

Join the early-access list. We'll walk your API through the full pipeline — import to signed release — on a demo project first, so you see the workflow before you commit.

  • Free guided walkthrough of your API (docs URL is enough)
  • Eval report on your connector candidate — keep it either way
  • Locked early-access pricing when we launch tiers

Request early access

We reply within two business days.

// opens your email client with everything pre-filled — no data leaves your machine until you hit send.

You're on the list.

Your email client should have opened with your request pre-filled — just hit send. We'll reply within two business days.