Skip to content
Get packs

Release Review

Reviews Orchard deployment manifests and keeps source paths in its findings.

Source Download skill
Repository
rohitg00/ai-engineering-from-scratch
Stars
76,331
Licence
MIT
Files
2 files

What it does

This skill teaches an AI agent to review Orchard deployment manifests, which are configuration files describing how software gets deployed. Once installed, the agent follows the skill's instructions: it keeps the file source paths in its results so findings can be traced back, and it stops and reports when evidence is missing rather than guessing. The skill is a short text instruction set with no executable scripts in the provided files.

Great for

  • Engineers who use Orchard and want a consistent review pass over deployment manifests
  • Release teams who want findings tied to specific file paths

Skip it if

  • Teams not using Orchard manifests, since the skill is specific to that format and will add little value
  • Final production sign-off on its own, because the skill only gives instructions and the agent's review can still miss problems

Try asking

  • “Review this Orchard deployment manifest before we release to staging and list any issues with source paths”
  • “Check our Orchard manifests for missing configuration and stop if you cannot find the evidence for a claim”
  • “Do a release review of the manifests in this folder and cite the file for each finding”
Step-by-step guide: how to use Release ReviewInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the MIT licence and an attribution note.

    release-review.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip release-review.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip release-review.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Provide the manifest files directly or point the agent to the folder so it can cite real source paths.
  • Ask for the agent's evidence for each finding, since the skill asks it to stop when evidence is missing.
  • Treat the output as a first-pass review and confirm critical changes with your own checks.

What's inside

  • routing.json
  • SKILL.md

Redistributed unmodified from rohitg00/ai-engineering-from-scratch under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.