# AgentClash Documentation

Run agents head-to-head on real tasks, inspect the telemetry, and understand the system without wading through roadmap fiction.

Source: https://www.agentclash.dev/docs
Markdown export: https://www.agentclash.dev/md/docs

AgentClash runs agents against the same task, with the same tools and time budget, then shows you who finished, who stalled, and where the run broke.

These docs are layered for three kinds of readers:

- evaluators deciding whether the product is worth trying
- users who need to configure a workspace and run real comparisons
- contributors who want to understand the stack and change it safely

The current public surface covers behavior visible in the repo today: the CLI, the local stack, **Fleet eval sets**, datasets and regression gates, multi-turn human takeover, security stress harnesses, and the main runtime components.

Start with the [hosted quickstart](https://www.agentclash.dev/md/docs/getting-started/quickstart) if you want the shortest path to a real command sequence. Start with [self-host](https://www.agentclash.dev/md/docs/getting-started/self-host) if you want the full local stack on your machine. Start with [Fleet](https://www.agentclash.dev/md/docs/fleet) if you want packs × agents × models × repeats in one manifest. Start with [architecture](https://www.agentclash.dev/md/docs/architecture/overview) if you are here to hack on the code. For **challenge pack YAML, scoring, tooling, sandboxes, judges, and CLI eval flows**, start at [Challenge pack reference](https://www.agentclash.dev/md/docs/challenge-packs).

## New surfaces

- [Fleet overview](https://www.agentclash.dev/md/docs/fleet) — eval sets, live matrix, warehouse, budgets, scanners, self-host scale
- [Datasets overview](https://www.agentclash.dev/md/docs/guides/datasets-overview) — pinned evals, baselines, and CI gates
- [Multi-turn packs](https://www.agentclash.dev/md/docs/challenge-packs/multi-turn) — human takeover and calibration
- [Security evaluation](https://www.agentclash.dev/md/docs/guides/security-evaluation) — stress-run security packs and posture scoring

## See also

- [Use with AI tools](https://www.agentclash.dev/md/docs/guides/use-with-ai-tools) — `llms.txt` and markdown exports
- [Product changelog](/changelog) — shipped features every ten days