Index of every page on this site: llms.txt
All skills
ntcoding avatar

/fix-it-never-work-around-it

@21c6101
by Nick Tunentcoding/claude-skillz366 stars
47
Run it yourself

Stops execution and fixes root cause when commands, builds, scripts, or tools fail unexpectedly. Triggers on workaround language: 'directly', 'instead', 'alternatively', 'skip', 'fall back', 'work around', 'isn't working', 'broken', 'manually'. Activates on any unexpected non-zero exit code or process failure.

Use this Skill: https://skilld.dev/gh/ntcoding/claude-skillz/fix-it-never-work-around-it

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈85 tokens for metadata: the name and description. ≈750 when used: this file.

Description uses 4.3% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈341 characters, or 4.3%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Fix It, Never Work Around It

Critical Rules

🚨 Rule 1: Follow the prescribed process exactly. If a step in the process fails, fix the root cause. Never skip, replace, or approximate the step. If you don't fix it, the problem recurs and all subsequent work is compromised. This does not apply to expected failures like TDD red phase — a failing test is the process working correctly.

Example: The code review subagent failed. ❌ "I'll review my own code and push." ✅ Fix why the subagent failed.

🚨 Rule 2: Follow instructions exactly. If a tool or command doesn't produce the expected result and there is no obvious alternative that produces the exact same result, stop immediately. Fix the root cause or ask for help. Never fabricate, approximate, or substitute.

Example: API authentication failed. ❌ "I'll define some likely values instead." ✅ Fix the authentication or ask the user.

Workaround Detection

If you catch yourself using any of these phrases, you are about to workaround. STOP.

Workaround Signal What You're Actually Doing
"Let me check via X directly" Bypassing the broken tool
"X is broken, let me use Y instead" Abandoning the fix
"Let me try a different approach" Avoiding the actual problem
"I'll work around this by..." Literally saying it
"Alternatively, we can..." Proposing a bypass
"Since X isn't working, let me..." Giving up on fixing X
"Let me just manually..." Bypassing automation
"I'll skip this step and..." Ignoring the failure
"Let me fall back to..." Retreating from the fix

Anti-patterns

❌ The Helpful Bypass

⏺ Bash(npm run build)
  ⎿ Error: Module not found '@company/shared'

⏺ The build has a dependency issue. Let me check the
  source files directly to understand the code structure.

The build is broken. Reading source files "directly" doesn't fix the build.

❌ The Tool Swap

⏺ Bash(nx show projects)
  ⎿ Error: Failed to process project graph

⏺ NX project graph is broken. Let me check the tasks
  via gh directly and read the PRD.

The NX project graph is broken. Using gh doesn't fix the project graph. Every NX command will fail until this is fixed.

Mandatory Checklist

When a command or process fails unexpectedly:

  1. Verify you have STOPPED your current task
  2. Verify you are investigating the ROOT CAUSE of the failure
  3. Verify your proposed fix makes the ORIGINAL command/process work
  4. Verify you are NOT using a different tool/source/approach to bypass
  5. Verify the original command succeeds after your fix

Do not resume your previous task until all checks pass.

🚨 REMEMBER: Following the process and following instructions is MORE important than achieving a result by any means necessary. Consistency and reliability are crucial. Do not improvise, do not try to be helpful when the process cannot be followed.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alerts1mo4 checks · Risk SAFE
  • Gen Agent Trust Hub1mo

    This skill provides instructions to ensure the AI agent addresses the root cause of technical failures rather than using workarounds. No security issues were detected.

  • Socket1mo

    No alerts

  • Snyk1mo

    Risk: LOW · No issues

  • Runlayer8mo

    1 file scanned · No issues

Signed by skilld at 21c6101. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Dormantupdated 10 months ago
version
1.0.0

README badge

README badge for ntcoding/claude-skillz/fix-it-never-work-around-it