Agent skill
repo-conventions-check
MANDATORY before adding files to any repo. Enforces checking existing patterns and confirming with the user before starting work. Use when: (1) adding new files, (2) adding new skills, (3) creating PRs, (4) any structural changes. BLOCKS progress until conventions are verified and user confirms the plan.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/data/repo-conventions-check
SKILL.md
Repo Conventions Check
MANDATORY before adding any files or making structural changes to a repository.
Blocking Condition
Before adding ANY new file or directory, you MUST:
- Find existing examples of similar files
- State the correct location based on those examples
- Tell the user your plan and wait for confirmation
- Send updates during work
If you skip any step: BLOCKED
Workflow
Step 1: Find Existing Patterns
Before adding a file, search for similar existing files:
# For skills
ls -la .claude/skills/ | head -10
find . -name "SKILL.md" -type f | head -10
# For any file type
find . -name "*.ts" -type f | head -10
find . -type d -name "similar-name" | head -10
Step 2: State the Location
Tell the user explicitly:
Based on existing skills in .claude/skills/, I will add the new skill to:
skills/my-new-skill/SKILL.md
NOT in packages/skills/skills/ (generated output; do not edit).
Step 3: Confirm Before Starting
DO NOT START WORK until you've told the user:
- What you're going to do
- Where files will go
- What the PR will contain
Wait for their confirmation or correction.
Step 4: Send Updates
During work, send updates every few minutes:
- What you just did
- What you're doing next
- Any issues or questions
Rationalizations (All Rejected)
| Excuse | Why It's Wrong | Required Action |
|---|---|---|
| "I know where it goes" | You were wrong last time | Check existing patterns |
| "It's obvious" | Clearly it wasn't | Confirm with user |
| "I'll tell them after" | Too late to fix mistakes | Tell them BEFORE |
| "They're busy" | Silence is worse | Brief update is fine |
| "It's a small change" | Small mistakes compound | Follow the process |
Examples
Wrong (BLOCKED)
User: Add a new skill for X
*immediately creates files without checking*
*puts skill in wrong location*
*doesn't tell user what's happening*
Correct
User: Add a new skill for X
Me: I'll add this skill. First let me check where existing skills live...
*runs: ls .claude/skills/*
Existing skills are in .claude/skills/. I'll create:
.claude/skills/x-skill/SKILL.md
Does that look right? I'll start once you confirm.
User: yes
Me: Starting now. I'll update you in a few minutes.
*does work*
Me: Done. Created the skill and pushed to branch feature/x-skill.
Creating PR now.
Enforcement
This skill chains with other skills:
- Before any file creation: repo-conventions-check MUST run
- Before any PR: verify locations match existing patterns
- During work: send updates via available channels
Self-Check
Before completing any task that adds files:
- Did I check existing patterns?
- Did I tell the user my plan?
- Did I wait for confirmation?
- Did I send updates during work?
If any answer is "no": STOP and fix it.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?