agentleFS
Sign inStart free
vibe-cokitvibe-cokit/claude-code/skills/vk-fix-hard/SKILL.md

vk-fix-hard

Skill1 starsChanged 4 months agoApache-2.0View on GitHub

⚡⚡⚡ Use subagents to plan and fix hard issues

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 1 is distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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
  1. Very specific:

    Use tester subagent to write test cases and run the tests, make sure it works, then report back to main agent.

    line 26
  2. Very specific:

    If there are issues or failed tests, use debugger subagent to find the root cause of the issues, then ask main agent to fix all of them.

    line 27
  3. Very specific:

    After finishing, delegate to code-reviewer subagent to review code.

    line 29
  4. Distinctive:

    If there are critical issues, ask main agent to improve the code and test everything again.

    line 30
  5. Very specific:

    Follow workflows/fix-report-contract.md — post-approval project-manager + docs-manager parallel wave, rejection path, Final Report, asset …

    line 35

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Fix Hard

Ultrathink to plan & start fixing these issues follow the Orchestration Protocol, Core Responsibilities, Subagents Team and Development Rules: <issues>$ARGUMENTS</issues>

Workflow

For video/screenshot input handling, see workflows/visual-input.md.

Break complex problems into sequential thought steps and state each hypothesis before testing it. Analyze the skills catalog and activate other skills that are needed for the task during the process.

  1. Parallel wave (per workflows/parallel-execution.md): debugger (root cause) + researcher (online lookup of the reported issue, if needed) run concurrently — researcher doesn't need debugger's output, only <issues>. Both report back to main agent.
  2. planner creates an implementation plan from BOTH reports (needs both → serial after the wave).
  3. Then use general agent (main agent) to implement the plan step by step.
  4. Use tester subagent to write test cases and run the tests, make sure it works, then report back to main agent.
  5. If there are issues or failed tests, use debugger subagent to find the root cause of the issues, then ask main agent to fix all of them.
  6. Repeat until all tests pass. Loop bound: after 3 fix→retest cycles still failing, STOP — the architecture/approach is likely wrong, not just the hypothesis. Re-question the design with debugger before any 4th attempt; never loop indefinitely.
  7. After finishing, delegate to code-reviewer subagent to review code. If there are critical issues, ask main agent to improve the code and test everything again. Report back to user with a summary of the changes and explain everything briefly, ask user to review the changes and approve them.

Closing

  1. Follow workflows/fix-report-contract.md — post-approval project-manager + docs-manager parallel wave, rejection path, Final Report, asset rules. Do not restate it here.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free