# Use with AI Tools

Feed AgentClash docs into ChatGPT, Codex, Claude Code, and similar tools using llms.txt and markdown exports, and install the AgentClash Agent Skills.

Source: https://www.agentclash.dev/docs/guides/use-with-ai-tools
Markdown export: https://www.agentclash.dev/md/docs/guides/use-with-ai-tools

Goal: give an assistant or coding agent enough AgentClash context to answer questions, draft workflows, or transform the docs into internal runbooks.

Prerequisites:

- You can open or paste URLs into the assistant you are using.
- You know whether you want the full docs bundle or just one page.

## Pick the right docs export

AgentClash exposes explicit content and contract surfaces for agents:

- `/llms.txt`: a compact index of the shipped docs set
- `/llms-full.txt`: one bundled markdown-oriented export of the full docs corpus
- `/md/...`: page-level Markdown exports that mirror every public page
- `/docs-md/...`: compatibility URLs for existing docs integrations
- `/openapi.yaml`, `/schemas/...`, and `/cli-schema.json`: API, artifact, and CLI contracts
- `/publications` and `/publications/sitemap.xml`: explicitly published, redacted evaluation artifacts

Use them differently:

- use `llms.txt` when the tool needs a map first
- use `llms-full.txt` when you want one-shot context for a larger prompt
- use `/md/...` when you only need one focused page, like quickstart or config
- use the machine contracts when an agent needs exact API, artifact, command, flag, or exit-code shapes

## Fastest workflow in ChatGPT, Codex, or Claude Code

1. Start with `https://www.agentclash.dev/llms.txt`.
2. If the tool can fetch URLs directly, give it that URL first.
3. If the tool cannot fetch URLs, open the file yourself and paste the contents.
4. Ask the tool which page it needs next.
5. Feed the relevant `/md/...` page or the full bundle, depending on scope.

That keeps the context tight. Do not dump the full bundle into every prompt by default.

## Good prompt patterns

Use prompts that ask the model to stay anchored to the supplied docs. Examples:

```text
Using https://www.agentclash.dev/llms.txt, tell me which docs pages I should read to self-host AgentClash and understand the worker architecture.
```

```text
Use the Markdown from https://www.agentclash.dev/md/docs/guides/interpret-results and turn it into a short incident-review checklist for my eval team.
```

```text
Use https://www.agentclash.dev/llms-full.txt as the product docs corpus and answer only from that material: what is the difference between a run, an eval, and a challenge pack?
```

## When to use page-level exports instead of the full bundle

Prefer page-level exports when:

- you are debugging one subsystem
- you want tighter answers with lower token cost
- the assistant tends to over-generalize when given too much context

Prefer the full bundle when:

- you want a holistic onboarding summary
- you are asking for a docs-wide rewrite or glossary
- you are building internal retrieval or indexing pipelines

## Verification

You should now be able to hand any of these URLs to a tool and get grounded answers:

- `https://www.agentclash.dev/llms.txt`
- `https://www.agentclash.dev/llms-full.txt`
- `https://www.agentclash.dev/md/docs/getting-started/quickstart`
- `https://www.agentclash.dev/openapi.yaml`
- `https://www.agentclash.dev/cli-schema.json`
- `https://www.agentclash.dev/schemas/prompt-eval.schema.json`
- `https://www.agentclash.dev/schemas/prompt-eval-result.schema.json`
- `https://www.agentclash.dev/schemas/voice-artifact-manifest.schema.json`
- `https://www.agentclash.dev/schemas/voice-live-continuity-report.schema.json`
- `https://www.agentclash.dev/schemas/voice-source-separation-report.schema.json`
- `https://www.agentclash.dev/schemas/voice-video-sync-report.schema.json`
- `https://www.agentclash.dev/publications`

## Troubleshooting

### The assistant cannot open URLs

Open the relevant endpoint yourself and paste the content directly.

### The answer is too vague

Use a narrower `/md/...` page instead of the full bundle.

### The answer mixes product claims with guesses

Tell the tool to answer only from the supplied docs export and cite the page title it used.

## Install AgentClash Agent Skills

Beyond feeding docs into a tool, AgentClash ships **Agent Skills** that teach a
coding agent to operate the CLI directly — setup, eval runs, scorecard reading,
CI gates, and challenge-pack authoring.

Install them with the AgentClash CLI. It is idempotent and writes only
`SKILL.md` files — never your `CLAUDE.md`, `AGENTS.md`, or `.mcp.json`.

Pick the line for the coding agent you use — each command installs the same
skills into that agent's skills directory. The trailing comment is where the
files land:

```bash
agentclash integration claude install    # Claude Code -> ~/.claude/skills/
agentclash integration codex install     # Codex       -> ~/.agents/skills/
agentclash integration cursor install    # Cursor      -> ~/.cursor/skills/
agentclash integration openclaw install  # OpenClaw    -> ~/.openclaw/skills/
agentclash integration hermes install    # Hermes      -> ~/.hermes/skills/
agentclash integration opencode install  # OpenCode    -> ~/.config/opencode/skills/
```

After installing, verify the integration (swap `claude` for your agent):

```bash
agentclash integration claude doctor      # report installed / missing / drifted skills
```

Project-local OpenCode installs use `.opencode/skills/`. There is no `--project`
flag; export straight into that directory instead:

```bash
agentclash skills export --dir ./.opencode/skills
```

Do not add `--host opencode` here: `--host` nests skills under the host's
home-relative path (for OpenCode, `.config/opencode/skills/`), which is right for
a home-directory install but would double-nest inside a project-local directory.
`--dir` alone writes the flat `<skill>/SKILL.md` layout.

### Offline export

For air-gapped machines, export the embedded snapshot to a directory or tarball,
then copy or extract into the agent skills path:

```bash
agentclash skills export --dir ./agentclash-skills
agentclash skills export --dir ./claude-skills.tar.gz --host claude --format tar.gz
tar -xzf claude-skills.tar.gz -C ~
agentclash integration claude doctor
```

Or install the published bundle with [GitHub CLI](https://cli.github.com) 2.90+,
which drops skills into the correct directory for Claude Code, Copilot, Cursor,
Codex, Gemini CLI, or Antigravity:

```bash
gh skill install agentclash/agentclash <skill>
```

The skills assume the `agentclash` CLI is on your PATH (`requires_cli: true`).
After installing, reload skills in your agent.

## See also

- [Quickstart](../getting-started/quickstart)
- [Config Reference](../reference/config)
- [Codebase Tour](../contributing/codebase-tour)
- [Datasets overview](https://www.agentclash.dev/md/docs/guides/datasets-overview)
- [Multi-turn packs](https://www.agentclash.dev/md/docs/challenge-packs/multi-turn)
- [Security evaluation](https://www.agentclash.dev/md/docs/guides/security-evaluation)