# ac.inference.sh/mcp

run any ai model. compose agents, stack knowledge, connect tools. one api, pay per run.

- **Type:** MCP server
- **Trust:** 60/100 (B), scored on the content rubric
- **Verification:** community-indexed — nobody has claimed this listing
- **Version:** 2.0.1
- **Author:** ac.inference.sh
- **License:** Unknown
- **Endpoints:** streamable-http https://api.inference.sh/mcp
- **Source:** https://api.inference.sh/mcp
- **Endpoint health:** reachable (last checked 2026-10-02T01:40:15.256Z, 12 samples) — uptime is not a security property and is not part of the trust score
- **Compatible clients:** claude-code, cursor, copilot, chatgpt, gemini (basis: transport)

## Trust

60/100 (B), scored on the content rubric
- Publisher verified: no
- Install scripts: nothing suspicious found
- Prompt-injection scan: clean
- Obfuscation scan: clean
- Evidence age: 45 days

## Security scan

**Assessment scope:** Hosted endpoint only: Forge can assess who publishes it and whether it answers, not the code behind it. No CVE, dependency or source analysis applies, so nothing here is a clean bill for that code.
- **Status:** probed (endpoint answered; no code examined)
- **Scanned:** 2026-08-28T12:14:14.094Z
- **Version scanned:** live
- **CVEs:** no coverage — this entry has no package coordinates to query OSV against, so "no known CVEs" is NOT asserted for it.

## Tools

25 declared. Observed from a live `tools/list` probe.
- `app_get` — Get detailed app info including input/output schemas and version history
- `app_list` — Search and filter available apps on inference.sh
- `app_run` — Run an app on inference.sh. Creates a task that executes the app with the given input.
- `knowledge_create` — Create a new knowledge entry (concept, observation, preference, reference, etc.)
- `knowledge_delete` — Delete a knowledge entry by ID
- `knowledge_get` — Get detailed information about a knowledge entry by namespace/name or ID
- `knowledge_list` — List your knowledge entries, optionally filtered by type
- `knowledge_search` — Search your knowledge entries by query
- `mcp_connect` — Initiate a connection to an MCP server. For OAuth servers, returns an authorization URL to open in a browser.
- `mcp_disconnect` — Remove connection to an MCP server
- `mcp_get` — Get details about an MCP server including connection status
- `mcp_list` — Browse available MCP servers that can be connected to
- `mcp_run` — Call a tool on a connected MCP server. The server must be connected first via mcp_connect.
- `mcp_search` — Search available MCP servers by query
- `mcp_tools` — List available tools on a connected MCP server, or get details for a specific tool
- `skill_files` — List supplementary files for a skill (references, scripts, etc.)
- `skill_search` — Search the public skill store
- `skill_store` — Browse approved skills in the public skill store
- `skill_upload` — Create or update a skill with SKILL.md content and optional supporting files
- `skill_use` — Fetch a skill's assembled SKILL.md content. Tries the inference.sh store first, falls back to GitHub. Returns the full skill content for the agent to consume.
- `skill_view` — View a specific supplementary file from a skill's manifest
- `task_cancel` — Cancel a running or queued task
- `task_get` — Get status and output of a task
- `task_list` — List your tasks on inference.sh
- `task_logs` — Get execution logs for a task

## Install

**Verdict: review** — Installable, but 3 things to check first: No CVE coverage: this entry has no npm/PyPI coordinates to query OSV against, so "no known vulnerabilities" is not a claim that can be made about it.
**Cautions** (coverage gaps and advisories — never blocking)
- No CVE coverage: this entry has no npm/PyPI coordinates to query OSV against, so "no known vulnerabilities" is not a claim that can be made about it.
- Hosted endpoint only: Forge can assess who publishes it and whether it answers, not the code behind it. No CVE, dependency or source analysis applies, so nothing here is a clean bill for that code.
- No publisher has proved control of this listing; it is indexed, not vouched for.
**Config** (claude-code):
```json
"{\n  \"mcpServers\": {\n    \"mcp\": {\n      \"type\": \"http\",\n      \"url\": \"https://api.inference.sh/mcp\"\n    }\n  }\n}"
```

## Blast radius

Moderate to extensive — no credential declaration found, from the publisher, the upstream registry, or the README. Known so far: deletes data; runs on someone else's infrastructure.
- Floor 33, ceiling 57 (tier: unknown)
- `unknown` means the floor and ceiling land in different bands — not measured enough to name one. It does not mean low.
- This is impact, not likelihood. A high radius is not a defect: a filesystem server is supposed to write files. It is never part of the trust score.

## Machine-readable views of this entry

- Signed JSON: https://forgeregistry.com/api/v1/packages/ac.inference.sh%2Fmcp
- Install plan: https://forgeregistry.com/api/v1/packages/ac.inference.sh%2Fmcp/install-plan
- Alternatives: https://forgeregistry.com/api/v1/alternatives/ac.inference.sh%2Fmcp
- HTML page: https://forgeregistry.com/registry/ac.inference.sh%2Fmcp
- MCP: POST https://forgeregistry.com/api/mcp → `forge_get_package` / `forge_install_plan`

## About this document

Generated by Forge (https://forgeregistry.com) — a compact rendering of the same record served, signed, at the JSON URL above. Trust and scan facts are the registry's own measurements; anything Forge did not measure is named as unmeasured rather than omitted.