# systemprompt.io

> AI governance infrastructure for organisations running agents at scale, delivered as a single self-hosted Rust binary. This file is for AI agents: it documents the free agent-callable tools API on this domain and how to use it.

## Free tools API

Machine-readable manifest with input schemas: https://systemprompt.io/api/v1/tools

Every request must carry an agent identifier: an `X-Agent-Id` header or an `agent_id` field in the JSON body (free-form, 1-64 chars). The `/run` endpoints are free and never ask for an email.

- **GitHub Actions permissions generator** — Minimal GITHUB_TOKEN permissions block from a list of workflow tasks, with fork and GitHub App caveats. Page: https://systemprompt.io/tools/github-actions-permissions — Endpoint: `POST https://systemprompt.io/api/v1/tools/github-actions-permissions/run`
- **CLAUDE.md scorer** — Scores a CLAUDE.md file against published best practices, returning a score out of 100 with per-rule recommendations. Page: https://systemprompt.io/tools/claude-md-scorer — Endpoint: `POST https://systemprompt.io/api/v1/tools/claude-md-scorer/run`
- **llms.txt and AGENTS.md generator and validator** — Generates or validates llms.txt and AGENTS.md files against the published rule catalogue. Page: https://systemprompt.io/tools/llms-txt-generator — Endpoint: `POST https://systemprompt.io/api/v1/tools/llms-txt/run`
- **EU AI Act risk classifier** — Annex III decision tree returning risk class, article references, obligations, and an evidence checklist. Page: https://systemprompt.io/tools/eu-ai-act-risk-classifier — Endpoint: `POST https://systemprompt.io/api/v1/tools/eu-ai-act/run`
- **AI coding tool ROI calculator** — Monthly cost, value, net ROI, break-even, and sensitivity range for an AI tooling rollout. Page: https://systemprompt.io/tools/ai-roi-calculator — Endpoint: `POST https://systemprompt.io/api/v1/tools/ai-roi-calculator/run`

Example call:

```bash
curl -s -X POST https://systemprompt.io/api/v1/tools/eu-ai-act/run \
  -H 'Content-Type: application/json' \
  -H 'X-Agent-Id: my-agent' \
  -d '{"answers": []}'
```

## Embedding these tools

Every tool has an embed page at `https://systemprompt.io/tools/{slug}/embed/` designed for iframes (width 100%, height 720). Humans and coding agents are welcome to embed any tool on their own sites and docs, provided the visible attribution link back to systemprompt.io below the iframe stays intact.


## Paid reports MCP server

One structured AI report tool, priced per call and settled by card (Stripe) or USDC on Base via the x402 protocol (no account, no API key, no subscription): EU AI Act compliance ($1.50). The deterministic Annex III decision tree grounds the AI analysis and its raw output ships verbatim in the report JSON. Discovery is free; call the tool unpaid to receive an HTTP 402 challenge with exact payment requirements. Endpoint: `https://systemprompt.io/api/v1/mcp/systemprompt-reports/mcp` (tool `eu_ai_act_compliance`). Full input schema, pipeline, and an example report: https://systemprompt.io/tools/eu-ai-act-compliance-report

## More

- Full site index for LLMs: https://systemprompt.io/llms.txt
- Discovery manifests: https://systemprompt.io/.well-known/agents.json and https://systemprompt.io/.well-known/ai-catalog.json
- Evaluation template: https://github.com/systempromptio/systemprompt-template
- Feedback or leads: open an issue labelled `feedback` on the GitHub repository, or email hello@systemprompt.io