postfact-mcp
MCP server that resolves ambiguous outcomes of side-effecting API/tool calls, returning DONE, NOT_DONE, or UNKNOWN with retry safety guidance.
README
POSTFACT
Did the call actually happen?
POSTFACT is a tiny ambiguous-outcome resolver for autonomous software. Use it after a side-effecting API/tool action fails ambiguously and before retrying.
Production: https://postfact-api-production.up.railway.app
It is intentionally strict:
DONEonly when affirmative evidence establishes completion.NOT_DONEonly when affirmative evidence establishes non-execution.UNKNOWNwhenever evidence is insufficient.UNKNOWNis the safe default.
POSTFACT does not execute the underlying action, move money, or pretend a timeout means failure.
REST
Production endpoint: POST https://postfact-api-production.up.railway.app/v1/resolve
Input accepts method/status/failure, side-effect and idempotency context, durable references, and optional affirmative evidence. Output is state, retry_safe, confidence, evidence, and next.
MCP
Remote endpoint: POST https://postfact-api-production.up.railway.app/mcp
Tool: resolve_outcome
The tool description explicitly tells agents to invoke POSTFACT after an ambiguous side-effect failure and before retry.
Middleware-shaped contract
const outcome = await postfact.resolve(failureContext)
if (outcome.state === 'DONE') return completed
if (outcome.state === 'NOT_DONE' && outcome.retry_safe) return retry()
return verifyOrHold()
Future SDKs can wrap retry middleware without changing the wire contract.
POSTFACT + RECOVER
POSTFACT answers what can we establish happened? RECOVER answers what is the safest next action? A future retry stack can call POSTFACT first, then RECOVER when the outcome remains uncertain.
Safe discovery and demo
- Health:
GET https://postfact-api-production.up.railway.app/health - Demo:
GET https://postfact-api-production.up.railway.app/demo - Activity:
GET https://postfact-api-production.up.railway.app/activity - Machine-readable activity:
GET https://postfact-api-production.up.railway.app/activity.json
GET /demo is deliberately excluded from real-use analytics.
Real-use scoreboard
GET /activity shows the permanent FIRST 10 TRUE STRANGERS scoreboard. GET /activity.json exposes its machine-readable summary. Demo/internal traffic is excluded. Production persistent storage is mounted at /data.
Run
npm test
npm start
Default port: 8080.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.