Metadata-Version: 2.4
Name: agentpress-static
Version: 0.2.0rc4
Summary: Make repos readable by AI agents with llms.txt, static discovery manifests, schema gates, proof receipts, and CLI checks.
Author: Barney Wohl
License: MIT
Keywords: agentpress,llms-txt,mcp,model-context-protocol,claude,gpt,agent-instructions,repo-onboarding,rag,evals,developer-tools,static-site
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Text Processing :: Markup
Classifier: Topic :: Internet :: WWW/HTTP :: Site Management
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Dynamic: license-file

# AgentPress

Make any repo readable by AI agents in 60 seconds with `llms.txt`, static discovery manifests, schema gates, proof receipts, and CLI checks.

## Compatibility

| Runtime | Support level | Notes |
|---------|---------------|-------|
| Claude (Anthropic) | Full | Native `llms.txt` plus agent contracts. |
| GPT-4 / Codex | Full | Uses `llms.txt` and machine-readable entrypoints. |
| Gemini | Full | Uses `llms.txt` and machine-readable entrypoints. |
| GLM-4 (Zhipu) | Full | Uses `llms.txt` and `.well-known/agentpress.json`. |
| LangChain | Full | Uses `llms.txt` and structured manifests. |
| CrewAI | Full | Uses `llms.txt` and structured manifests. |
| MCP / ACP | Full | Uses `.well-known` manifests and agent contracts. |
| Cline / Roo / OpenHands | Compatible | Uses `llms.txt` as the first-read handoff. |

## Execute

```bash
python3 scripts/agentpress.py doctor --json
python3 scripts/agentpress.py list --json
python3 scripts/agentpress.py validate agentpress/examples/api-docs-handoff --json
python3 scripts/agentpress.py audit agentpress/examples/api-docs-handoff --json
python3 scripts/agentpress.py eval agentpress/examples
python3 scripts/agentpress.py check-registry
python3 scripts/agentpress.py check-openapi
```

## First-user eval harness

Run a safe local adversarial smoke test against synthetic repos:

```bash
python3 scripts/eval_first_user_harness.py --timeout 20
```

The harness creates 5-10 generated repos under `/tmp`, runs local AgentPress first-contact commands (`--help`, `doctor --mode local`, `lint --no-write`, `first-run-wizard --no-write`, `first-user-bootstrap --no-write`), writes a Markdown backlog report, and cleans generated repos by default.

## Required agent files

- `AGENT_ENTRYPOINT.md`
- `agent-task-card.json`
- `source-map.json`
- `freshness.json`
- `allowed-actions.json`
- `.well-known/ai-ingestion.json`
- `llms.txt`
- `sitemap.xml`
- `citation-policy.md`
- `disclaimer.md`
- `evals/*.jsonl`

## Reference bundles

- API/docs agent handoff → [`examples/api-docs-handoff`](./examples/api-docs-handoff/)
- Incident/runbook agent handoff → [`examples/incident-runbook-sharing`](./examples/incident-runbook-sharing/)
- Dataset/RAG agent handoff → [`examples/dataset-card-reuse`](./examples/dataset-card-reuse/)
- Knowledge transfer → [`examples/agent-knowledge-sharing`](./examples/agent-knowledge-sharing/)
- Agent compatibility → [`examples/universal-agent-reachability`](./examples/universal-agent-reachability/)

## Machine contracts

- [`agentpress-registry.json`](./agentpress-registry.json)
- [`articles/article-index.json`](./articles/article-index.json)
- [`protocols/mcp-manifest.json`](./protocols/mcp-manifest.json)
- [`protocols/executable-contracts.json`](./protocols/executable-contracts.json)
- [`schemas/`](./schemas/)
- [`self-test/self-test.json`](./self-test/self-test.json)
