Skip to content

Post-Release Babysitter

Watches a shipped claude-mem release for regressions across packages, hosted service and installed plugins.

Stars
110,174
Licence
Apache-2.0
Files
1 file

What it does

After a release is merged or published, this skill tells the agent to keep checking that what users actually receive works: the published package, the live production service, and the plugin installed on a user's machine. It also watches CI results and late user reports for a set window, verifies any problem before changing code, and proposes or makes focused fixes through the normal release process. It is written around the claude-mem project and expects its tooling and conventions.

Great for

  • Maintainers of claude-mem or similar plugins who ship packages and hosted services and want post-release verification
  • Release engineers who want a structured checklist for the hours after a deployment
  • Teams that already use babysit and version-bump style skills in the same repository

Skip it if

  • Projects with no release pipeline, since most checks assume a package registry, a hosted alias and a plugin cache (the skill would be mostly inapplicable)
  • Anyone expecting it to authorize rollbacks, new releases or announcements on its own; the file says monitoring does not grant that permission, but an agent could still overreach if the user is careless
  • Pre-release PR readiness work, which the description says belongs to the babysit skill instead

Try asking

  • “Watch the release we just merged for version 2.x and tell me if the npm package and production service look healthy”
  • “Check whether the published package has its provenance attestation and that the dist-tag points to the new version”
  • “Verify the installed claude-mem plugin is running the same version as the enabled cache after the upgrade”
  • “Review new GitHub issues and late CI failures since the release and summarise which ones are real regressions”
Step-by-step guide: how to use Post-Release BabysitterInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the Apache-2.0 licence and an attribution note.

    childhood.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

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

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip childhood.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

  • Give it the release version and an explicit observation window so it does not fall back to its 30-minute default
  • Provide credentials through your environment rather than pasting them in chat, since the skill says never to print them
  • Read the verification table first and decide which surfaces actually shipped; the agent will label anything it could not exercise as unverified

What's inside

  • SKILL.md

Redistributed unmodified from thedotmack/claude-mem under the Apache-2.0 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.