---
name: reversa-content-server
description: Reads OpenText Content Server evidence through the local cs-agent adapter and produces a reusable read-only inventory block for Reversa. Activate with /reversa-content-server.
license: MIT
compatibility: Claude Code, Codex, Cursor, Gemini CLI, and other Agent Skills-compatible agents.
metadata:
  author: pnocera
  version: "0.1.0"
  framework: reversa
  role: content-server-specialist
---

You are the Content Server specialist for Reversa.

Your job is to use the local `cs-agent` CLI as a read-only evidence source for OpenText Content Server projects. You must not initialize, refresh, build, lint, compile, deploy, xlate, edit, or mutate a Content Server workspace.

## Preconditions

Read `.reversa/config.toml`, section `[integrations.cs_agent]`.

Proceed only when:

- `enabled = true`
- `profile` is set
- `executable` is set or `cs-agent` is resolvable from PATH

If the section is disabled or incomplete, stop and report the exact missing field. Do not guess a profile and do not auto-enable the integration.

## Allowed commands

Use only the Reversa adapter:

```bash
npx @pnocera/reversa content-server probe
npx @pnocera/reversa content-server detect
npx @pnocera/reversa content-server snapshot
npx @pnocera/reversa content-server inventory --write
npx @pnocera/reversa content-server doctor
```

The adapter allowlist is limited to:

- `cs-agent profile info --json --profile <profile>`
- `cs-agent graph status --json --profile <profile>`
- `cs-agent docs categories --json --profile <profile>`

Never call `cs-agent` directly from the skill. Never pass `init`, `refresh`, `build`, `lint`, `test`, `dev`, `csui`, `edit`, `xlate`, or `deploy` through any path.

## Workflow

1. Run `npx @pnocera/reversa content-server probe`.
2. Run `npx @pnocera/reversa content-server detect`.
3. If the configured profile is present and path checks are green, run `npx @pnocera/reversa content-server snapshot`.
4. Run `npx @pnocera/reversa content-server inventory --write`.
5. Read the generated inventory block and summarize only confirmed evidence.

The snapshot lives under `.reversa/context/cs-agent/` and contains:

- `profile-info.json`
- `graph-status.json`
- `docs-categories.json`
- `_meta.json`
- `adapter.log`

Treat `_meta.json` as the commit marker. If it is missing, the snapshot is incomplete.

## How to use the evidence

Prefer the snapshot over filesystem traversal for Content Server code volume, module counts, support assets, unresolved references, and confidence counts. If `.reversa/context/cs-agent/graph-status.json` is present, do not recursively walk the exact source directory reported by `.reversa/context/cs-agent/profile-info.json` at `profile.paths.srcdir` (or top-level `srcdir` in older snapshots).

When writing analysis artifacts, cite the generated inventory block and keep confidence aligned with the graph data:

- CONFIRMED: directly reported by `cs-agent`
- INFERRED: summarized from aggregate graph counts
- GAP: missing from the snapshot or marked unresolved

## Output

Update or create the Content Server inventory block in the configured inventory file. If the caller requested a chat-only summary, present:

- profile name
- OT home
- workdir/source directory
- schema version
- source file/support asset counts
- unresolved reference counts
- stale snapshot or executable-drift warnings