before-edit
Skill18 starsChanged 5 months agoMITView on GitHub
Run before modifying a function — impact + test-map + explain in one pass, presented as a modification checklist.
In this repository22 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- archeoSkill
- auditSkill
- boundary-probeSkill
- check-my-workSkill
- cqs-batchSkill
- cqs-bootstrapSkill
- cqs-planSkill
- cqsSkill
- cqs-verifySkill
- docs-reviewSkill
- groom-notesSkill
- idleSkill
- investigateSkill
- landSkill
- migrateSkill
- prSkill
- recall-gateSkill
- reindexSkill
- releaseSkill
- troubleshootSkill
- update-tearsSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Run before modifying any function.
- Very specific: line 17
Run cqs impact <function_name> --json via Bash
- Very specific: line 48
If cqs impact returns empty results, the function may be dead code or a leaf node — note this
- Very specific: line 49
If cqs test-map returns empty, flag as "NO TESTS — write tests before and after editing"
- Very specific: line 50
If cqs explain fails (function not indexed), fall back to cqs "<function_name>" --name-only --json
- Distinctive: line 51
Always run all 3 commands before presenting — don't skip based on partial results
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Before Edit
Run before modifying any function. Shows what breaks, what tests cover it, and what to check.
Arguments
<function_name>— the function you're about to edit
Process
- Run
cqs impact <function_name> --jsonvia Bash - Run
cqs test-map <function_name> --jsonvia Bash - Run
cqs explain <function_name> --jsonvia Bash - Present results as a checklist
Output Format
Present this to the user (fill in from the JSON results):
## Impact Analysis: <function_name>
**Signature:** <from explain>
**File:** <from explain>
**Callers (<count>):** <list top 10 caller names with files>
**Tests (<count>):** <list test names>
**Risk:** <High if >10 callers, Medium if 3-10, Low if 0-2>
### Before you edit:
- [ ] Understand what callers expect (check the top callers above)
- [ ] Note the existing tests — run them after your change
- [ ] If changing the signature or return type, update all callers
### After you edit:
- [ ] Run: `cargo test --features cuda-index -- <test_names>`
- [ ] If tests fail, check whether your change broke the caller contract
- [ ] If no tests cover the changed behavior, write oneRules
- If
cqs impactreturns empty results, the function may be dead code or a leaf node — note this - If
cqs test-mapreturns empty, flag as "NO TESTS — write tests before and after editing" - If
cqs explainfails (function not indexed), fall back tocqs "<function_name>" --name-only --json - Always run all 3 commands before presenting — don't skip based on partial results
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

