Skip to main content
skill CX Codex

childhood — Claude Code skill: what it does, how to install

The childhood skill stays with a release after publication or deployment, checking shipped artifacts, production behavior, plugins, and late CI signals.

Apache-2.0 570 installs ★ 118,868 updated 2026-11-16

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add thedotmack/claude-mem --skill childhood -a claude-code
Open on GitHub

What childhood does

childhood keeps release verification running after a merge, publish, or deployment. It checks that the tag points to the shipped commit, reviews CI and late reports, and inspects the exact registry version, distribution tag, hash, entry points, and downloaded files. For hosted services it checks the live deployment, the changed user path, relevant errors, and migration state. For plugins it compares the enabled install, cache, worker process, and hook configuration.

The skill follows an observation window, polls changing checks, and separates regressions from pre-existing issues or infrastructure failures. A confirmed problem is handled through a focused fix and the repository’s normal PR, release, or deployment path, followed by another pass after the repair ships. Monitoring alone does not authorize a release, rollback, production migration, or public announcement.

For a claude-mem release, it exercises Smart search through separate search, timeline, and get_observations calls, plus optional dumb_search when available. It also checks provenance attestation. An unavailable API leaves a coverage gap rather than proving the release is healthy.

What it helps with

  • verify that a published package contains the required files
  • match a live deployment to the shipped commit
  • catch late CI failures and actionable user reports
  • inspect an installed plugin and its running worker
  • follow a verified regression through a repair release

Example prompts for the agent

  • Use the childhood skill to check the published package and production after the release.
  • Watch the latest deployment and review CI, the changed user path, and reports that arrived later.
  • Inspect the installed claude-mem plugin, its cache, worker process, and hook configuration.

How to install childhood

1Skills CLI (Vercel)

npx skills add thedotmack/claude-mem --skill childhood -a claude-code

2Manually: put the files in place

~/.claude/skills/childhood/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/childhood/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about childhood

How do I install childhood in Claude Code?

Run npx skills add thedotmack/claude-mem --skill childhood -a claude-code, then open ~/.claude/skills/childhood/SKILL.md.

What does childhood do?

It stays with a shipped release after merge, publication, or deployment. The skill checks delivered artifacts, production behavior, and installed plugins, then routes verified regressions through the normal repair process.

What happens when a release check cannot finish?

The release is reported as partially verified with the coverage gap and next action. An unavailable API is treated as missing coverage, not as evidence that the release is healthy.

Source text

SKILL.md · Apache-2.0
---
name: childhood
description: Babysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.
---

# Childhood

Stay with a release after it ships. Verify what users receive and what production serves, then fix verified regressions within the user's authorized scope. A successful merge or publish command is the start of this work.

## Establish the release

- Identify the repository, release version, immutable tag commit, merge commit, production deployment, and affected delivery surfaces. Use the release receipt and actual provider/registry state; do not guess from the current branch or today's latest build.
- Record the release, observation start/end, checks, findings, and next action with `work_state_write` when available. Keep announcements and unrelated infrastructure failures distinguishable from product regressions.
- Carry forward existing authorization. Monitoring does not itself authorize another release, rollback, production migration, or external announcement. Prepare and validate a concrete fix before requesting any authorization still missing.

## Verify the delivered release

Choose the checks relevant to its delivery surfaces; independent checks can run in parallel.

| Surface | Evidence to collect |
| --- | --- |
| Source and CI | Confirm the tag resolves to the shipped commit. Inspect CI for that commit and any release/changelog commits, including late failures. Paginate new issue reports, merged-PR comments, reviews, and review threads since release; verify actionable reports against the shipped code. |
| Published package | Check the exact registry version and the intended distribution tag separately. Verify package hash, entry points, and required files in the downloaded artifact; run a small installed CLI smoke from an isolated directory. Preserve the command's actual exit code when filtering output. |
| Hosted service | Confirm the live production alias serves the intended deployment/commit. Check health and the changed user path through the real authenticated interface, plus relevant error signals. For schema changes, verify the normal migration ledger reports no pending migrations or drift. |
| Installed plugin | Discover the user's actual enabled registration/cache, not just the marketplace copy. Compare changed source/distribution artifacts and open a fresh tool connection. Compare the running worker's version/path to the enabled install; an updated cache can still talk to a stale process. Exercise the affected install/upgrade handoff where authorized. Check hooks or watcher configuration in their real project/runtime scope; label behavior you did not exercise as unverified. |

For a claude-mem search release, exercise Smart search as three separate calls: search for an index, timeline for context, and get_observations for only the selected records. Exercise optional dumb_search explicitly where available, and readable invalid-input guidance through fresh local and hosted MCP connections when those surfaces shipped. Verify that Smart is the default, no auto or guided mode or continuation cursor is exposed, and tool results are concise text without pure JSON or duplicated structured metadata. Keep probes read-only unless writes are part of the explicit test scope. Never print credentials or private result bodies.

Publication can lag behind a successful command. An exact-version 404 or stale distribution tag remains pending until the registry agrees; use bounded retries and investigate the actual publish target/status before republishing. A warning alone is not a regression: verify the final artifact. When the repository publishes through CI, a failed publish job is a finding to fix through the normal release path, not to cover with a manual publish; for claude-mem, also confirm the published version carries a provenance attestation.

## Observe and repair

- Honor the requested observation window. Without one, watch through 30 minutes after the release first becomes publicly available or production becomes ready, whichever affected surface is later. For an older release, do a fresh sweep and inspect the completed late CI and reports covering that interval.
- Poll pending or changing checks every 30–60 seconds; check steady signals every 2–5 minutes. Keep individual waits at most 60 seconds so user steering and progress updates remain responsive. An unavailable API is a coverage gap, not evidence that the release is healthy.
- Verify each finding before changing code. Distinguish a release regression, a pre-existing issue, an operational failure, and a resolved/transient signal; give each its own evidence and next action.
- For a verified claude-mem release regression, pull back the promotion with the **Pull back a bad release** bullet in [version-bump](../version-bump/SKILL.md) step 13. A pull-back downgrades no install and is part of the release already authorized, so it needs no new authorization. Then ship the fix as a new patch release, as the next bullet describes.
- Make focused fixes, run checks appropriate to the change, and use the repository's normal PR/release/deployment process. Use [babysit](../babysit/SKILL.md) for a repair PR and [version-bump](../version-bump/SKILL.md) when a new claude-mem release is authorized. Published versions and existing tags stay immutable.
- After a repair ships, restart observation for the changed delivery surface and rerun its affected user path. Preserve unrelated worktrees, workers, registrations, user settings, and inference/provider policy.
- A repeated infrastructure failure or deleted notification webhook needs a concrete follow-up. Keep it pending with its cause and owner action; do not repeatedly retry it or announce success to another destination.

## Finish with evidence

Finish when the relevant CI is settled, delivered artifacts and affected user paths pass, no actionable release regression remains, and the observation window is satisfied. Do one final fresh sweep before reporting.

If a required check cannot be completed, report the release as partially verified with that coverage gap and next action. Separate an externally blocked operational follow-up from verified product health; do not silently mark either as resolved.

Return concise purpose-specific text: release/tag/deployment identity, checks and observation times, verified findings and repairs, and anything still pending. Keep raw API envelopes, internal metadata, and complete logs out of model-facing output. Link to a compact receipt when useful.

© thedotmack · Apache-2.0 · thedotmack/claude-mem

Similar skills

All skills for DevOps and infrastructure →

From the same set30

The whole set →
ccs-align ccs-align — a skill to manage the local claude-mem worker: health check, pull observations into cache, and detect rule conflicts (SHADOW_HOUSE, DENY_ALLOW, DRIFT, CLOCK_HEADER). Runs hourly. CX Codex +1 1,000installs Agent Cost Report Generates agent cost reports over chosen periods. Runs during session audits to measure outcome costs and wasted spending. CX Codex +1 816installs handoff Generates a structured HANDOFF.md with goals, failed attempts, and next steps. Use it when context degrades or when starting a fresh session. CX Codex +1 804installs screenpipe Searches local screen and audio logs captured by Screenpipe and saves selected findings into claude-mem. Useful when you need to recall meeting notes or app windows. CX Codex +1 660installs request-progress request-progress — a skill for Claude Code that groups recent conversations into overall requests and tracks build velocity from memory. Activates on requests to refresh progress or review request status. CX Codex +1 469installs owner-goals The owner-goals Claude Code skill maps an owner’s goals from prompts they wrote themselves. It runs on requests to review personal goals or refresh Goals in My Work. CX Codex +1 468installs backup-setup The backup-setup skill runs on Claude Code and manages local database snapshots for claude-mem. Sets up automatic backups, checks status, and restores databases. CX Codex +1 435installs community-sweep The community-sweep Claude Code skill reviews open pull requests and issues in claude-mem. Run it to triage community contributions, help finish PRs, or handle new issues. CX Codex +1 192installs

Project sponsors

Companies that keep this analytics open