Surgical Patch
Surgical-patch is a short instruction set for bug fixes and small behavior changes that keeps Claude Code from wandering. It tells the agent to reproduce the failure first when that's cheap, trace the symptom to the mechanism responsible, and change only the narrowest layer that owns the wrong behavior. It also forbids drive-by cleanup, renames, and new abstractions, and asks for only the regression proof the task needs. Reach for it when you want a tight diff you can review in one sitting rather than a helpful refactor you didn't ask for. The honest caveat: the rules are prose, not enforcement, so you still have to check the diff, and on a messy codebase "narrowest layer" can take some judgment to apply well.
- skills.sh installs
- 119.5K
- GitHub stars
- 118.7K
- last updated
- Oct 26, 2026
Install
$ npx -y skills add juliusbrussee/caveman --skill surgical-patch --agent claude-codeInstalls into .claude/skills of the current project.
Files
Surgical patch
Reproduce failure first when economical; otherwise capture strongest available evidence.
- Trace symptom to responsible mechanism.
- Change narrowest layer that owns incorrect behavior.
- Preserve unrelated behavior and user changes.
- Avoid cleanup, renaming, and abstraction outside fix.
- Add only regression proof relevant to task.
Run focused proof plus nearest affected gate. Stop when failure is fixed and regression proof passes.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageDetails
- Repository
- github.com/juliusbrussee/caveman
- Category
- Debugging
- First seen
- Oct 26, 2026
More from juliusbrussee/caveman
All 25- Lean Build119.1K
- Caveman Evidence Review119K
- Caveman Discover119K
- Caveman Manage118.7K
- Migration118.4K
- Caveman Setup118K
- Compress45.8K
- Ultracave41.3K
- Megacave40.1K
- Caveman Es513
- Caveman Cn437
- Caveman618.3K
- Caveman Commit431.6K
- Caveman Compress428.9K
- Caveman Review427.5K
- Caveman Help418.2K
- Cavecrew374.2K
- Caveman Stats374.1K
- Caveman Explore122.4K
- Caveman Learn121.5K
- Safe Refactor121.2K
- Caveman Optimize120.5K
- Investigate First120.1K
- Verify And Stop119.6K
Recommended
More Debugging
solana-devsolana-foundation/solana-dev-skillUse when user asks to "build a Solana dapp", "write an Anchor program", "create a token", "debug Solana errors", "set up wallet connection", "test my Solana pro108.9Kskills.sh installs
microsoft-foundrymicrosoft/github-copilot-for-azureDeploy, evaluate, fine-tune, and manage Foundry agents end-to-end: Docker build, ACR push, hosted/prompt agent create, batch eval, continuous eval, prompt optim103.1Kskills.sh installs
nx-workspacenrwl/nx-ai-agents-configExplore and understand Nx workspaces. USE WHEN answering questions about the workspace, projects, or tasks. ALSO USE WHEN an nx command fails or you need to che88.7Kskills.sh installs
auditpbakaus/impeccableRun technical quality checks across accessibility, performance, theming, responsive design, and anti-patterns. Generates a scored report with P0-P3 severity rat81.9Kskills.sh installs
test-driven-developmentaddyosmani/agent-skillsDrives development with tests. Use when implementing any logic, fixing any bug, or changing any behavior. Use when you need to prove that code works, when a bug71.3Kskills.sh installsdoubt-driven-developmentaddyosmani/agent-skillsSubjects every non-trivial decision to a fresh-context adversarial review before it stands. Use when correctness matters more than speed, when working in unfami65.8Kskills.sh installs