- Home
- Skills
- Agents & Orchestration
- repoctx+
Works with the AI Agents you already use
repoctx+
Fleet-scale context enforcement and stack-specific playbooks for repoctx documentation workflows.
$4.99
repoctx+
Example session with this skill installed
Run a fleet audit on my services workspace and show me the health score for service-a and service-b.
- Read your context and instructions
- Compiled the repoctx+
I have completed the fleet-wide context audit for your workspace. Below is the health report based on the repoctx engine analysis.
Fleet context health
Score: 50/100 · Repos: 2 · Failing: 1
| Repo | Score | Failing check | Exact finding |
|---|---|---|---|
| service-a | 50/100 | commands | npm run lint — ghost command, no longer present |
| service-b | 100/100 | — | — |
What to fix first
Regenerate service-a: The audit found a ghost command in the manifest. Run the generator to sync the tables with the current package.json state.
2. Verify paths: Once regenerated, perform a final audit to ensure no stale backticked paths remain in your prose.
Next steps
- Run
repoctx generate .inside theservice-adirectory to clear the ghost command. - Execute
repoctx audit . --checklocally to verify the fix before pushing to CI. - Review
references/playbooks/saas.mdfor tips on managing environment templates in service fleets.
Connects securely to your tools. The creator never sees your data.
About this skill
Managing context truth across a growing microservice or monorepo fleet is a losing battle. Documentation drifts the moment a developer renames a script or moves a file, leaving AI agents with hallucinations and broken paths.
This skill provides a team-scale workflow for repoctx to enforce context accuracy across three or more repositories. It delivers pre-tuned stack playbooks, versioned policies, and fleet-wide audit tools to ensure your AGENTS.md and CLAUDE.md files never lie to your LLMs.
What it does
- Fleet-wide auditing scans multiple repositories to generate drift reports and health scores.
- Stack-specific playbooks provide expert guidance on detection blind spots for Java, SaaS, Game-Dev, and Monorepos.
- Automated bootstrapping installs project-specific
repoctx.tomlpolicies andAGENTS.mdtemplates. - CI enforcement gates builds by exiting with an error if any repository context has drifted from the source code.
- Incremental gating allows teams to adopt context enforcement on new changes only using the
--sinceflag.
How it works
- Bootstrap use
scripts/setup.shto apply a stack-specific policy and template to a new or existing repository. - Synchronize run
repoctx generateto populate code-derived tables while preserving human-written prose. - Audit execute
scripts/fleet-audit.shat the workspace root to check every repository against the fleet policy. - Enforce integrate
repoctx audit --checkinto your CI pipeline to prevent stale documentation from being committed.
Frameworks & tools
Requires repoctx CLI v0.10.0 or later. Supports Java (Maven/Gradle), SaaS, monorepo, microservices, and game engines like Unity, Godot, and Unreal.
Why this beats prompting it yourself
Prompting an agent to audit context is unreliable because the agent relies on the very context that might be stale. This skill uses the repoctx engine to perform hard manifest-to-file checks, catching ghost commands and broken paths that manual prompting misses.
Use cases
- Enforce a single context policy across 10+ microservices.
- Generate stack-aware
AGENTS.mdfiles for a new Unity game project. - Identify "ghost commands" in documentation after a major package.json refactor.
- Prevent CI merges when a developer renames a file without updating the context link.
Known limitations
Does not support editing human-written prose outside of defined repoctx markers. Requires a repoctx+ license and CLI installation. Fleet mode requires a specific roots configuration in the workspace root.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.
Requires repoctx CLI v0.10.0+. Cannot edit human prose sections. Fleet mode expansion requires a repoctx.toml at the workspace root. Does not include the MCP server features.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 25 days ago
- Software
- @sriruma/repoctx
Node.js