# Library Guides


<div class="grid">

<a href="/docs/libraries/agent-evaluations/">

### Agent Evaluations

Run agent-as-judge evaluations in CI with `fit-eval supervise` and inspect the
resulting traces with `fit-trace`.

</a>

<a href="/docs/libraries/agent-collaboration/">

### Agent Collaboration

Run multi-agent sessions with `fit-eval facilitate` — specialists coordinate via
`Ask`, `Share`, and `Tell`, captured as a trace for `fit-trace`.

</a>

<a href="/docs/libraries/trace-analysis/">

### Trace Analysis

Read agent execution traces with `fit-trace` as qualitative research — orient,
code observations, find patterns, and write grounded findings.

</a>

<a href="/docs/libraries/xmr-analysis/">

### XmR Analysis

Distinguish stable processes from special causes with `fit-xmr` — record
observations over time, read signal rules, and turn variation into action.

</a>

</div>

Looking for product workflows like authoring standards, agent teams, or landmark
quickstart? See [Product Guides](/docs/products/).