After Change Monitor — Development skill for Claude Code
After making any non-trivial change, define a verification window and check back.
How to install After Change Monitor
Installs to ~/.claude/skills/tkwong-claude-ops-after-change-monitor/SKILL.md
mkdir -p ~/.claude/skills/tkwong-claude-ops-after-change-monitor && curl -fsSL https://raw.githubusercontent.com/tkwong/claude-ops/HEAD/skills/after-change-monitor.md -o ~/.claude/skills/tkwong-claude-ops-after-change-monitor/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What After Change Monitor does
name: after-change-monitor description: After making any non-trivial change, define a verification window and check back. Don't fire-and-forget.
after-change-monitor
You just changed something — config, code, restart, parameter. **Now you need to verify it actually had the effect you intended.** Don't move on to the next task until you've checked back at least once.
The mindset
A change is a *hypothesis*: "I think doing X will cause Y." You verify by measuring Y before and after
Alternatives in Development
- Contributors — Thank you to everyone who has contributed to making planning-with-files better 16.8k ★
- Skill Decision Support — Present options with trade-offs for informed decision-making — use when choosing between approaches 2.8k ★
- Backpass — You don't write AGENTS.md 1.1k ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Pre Ship Adversary
Before committing a non-trivial change, spend 2-3 minutes attacking the change yourself — scope creep, failure
Prompt Brief
An agent skill that writes a brief before the work: subagents map the code, then the agent drafts a paste-back
Arch Decision
Architecture Decision Record (ADR) generator backed by the Claude Power Pack vault. Before committing to a non
Mastering Java
I have a goal to do one leet code type problem every day, starting very simple and getting 1% harder each day
Repurpose Skill
See a reel with a fire hook? Don't start from scratch. Drop the link, and Claude downloads it, transcribes it,
Verification Gate
Use before claiming any work is complete, fixed, or passing - requires running verification commands and showi
Related Agents
General Skills Executor
Generic isolated runner for noisy "fire-and-forget" skills whose raw output would pollute the main context — c
History Reviewer
Git history-aware reviewer. Use when changes touch code with non-trivial history, or to check whether a fix ac
Paid Traffic Tracking
Use to implement or fix conversion tracking - GTM, GA4, Meta Pixel/CAPI, Google enhanced conversions, offline