agentleFS
rajasdevelrajasdevel/outcomebound/.agents/skills/diagnose/SKILL.md

diagnose

Skill0 starsChanged 15 days agoApache-2.0View on GitHub

Use when a failure's cause is unknown, a fix did not hold, or a defect is reported for repair. A declared stub failing as declared is no defect; handle the requested work directly.

195 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use zoom-out, to-prd and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • victorgarciaesgiSkill

    zoom-out

    Tell the agent to zoom out and give broader context or a higher-level perspective. Use when you're unfamiliar with a section of code or need to understand how it fits into the bigger picture.

    Also in 61 projects that use this

  • GoogleSkill

    to-prd

    Turn the current conversation into a PRD and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.

    Also in 56 projects that use this

  • GoogleSkill

    to-issues

    Break a plan, spec, or PRD into independently-grabbable issues on the project issue tracker using tracer-bullet vertical slices.

    Also in 53 projects that use this

  • simo97Skill

    edit-article

    The user will invoke this skill to help them edit an article.

    Also in 17 projects that use this

  • simo97Skill

    write-a-skill

    Create new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.

    Also in 44 projects that use this

  • AnasX7Skill

    setup-pre-commit

    Set up Husky pre-commit hooks with lint-staged (Prettier), type checking, and tests in the current repo. Use when user wants to add pre-commit hooks, set up Husky, configure lint-staged, or add commit-time formatting/typechecking/testing.

    Also in 27 projects that use this

  • +4 more from simo97

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 176 edited versions we read keep these.

  • 7

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the error makes the cause plain, fix and check it without a cause search.

    line 10
  2. Distinctive:

    If asked only what is wrong, answer from reads and checks without changes, and say what did not run.

    line 11
  3. Distinctive:

    Before changing anything, reproduce the reported input with its smallest failing test or calling command.

    line 15
  4. Distinctive:

    If reproduction fails, report what you tried.

    line 16
  5. Distinctive:

    Read the whole error, its first failure and named line.

    line 18
  6. Distinctive:

    When the error and reported case do not settle narrowing, read narrowing techniques.

    line 22
  7. Distinctive:

    If a fix fails with no new evidence, return to observations and list and check unobserved assumptions.

    line 33

“Common” is measured: kept by at least 6 in 10 of the 176 edited versions of this file we could read. “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

Diagnose a failure

Done means the failure is gone because its cause is: reproduce it or report it not reproduced, state the cause and its evidence, remove that cause, and report what checked the fix. If the error makes the cause plain, fix and check it without a cause search; still do the sweep below. If asked only what is wrong, answer from reads and checks without changes, and say what did not run.

Reproduce and find the cause

Before changing anything, reproduce the reported input with its smallest failing test or calling command; neighbouring or passing cases do not reproduce it. If reproduction fails, report what you tried; a fix without reproduction stays unverified. For a flaky failure, choose runs suited to its frequency and report runs and failures; if none fails, uncertainty remains; after the fix, the same number of failure-free runs is its evidence. Read the whole error, its first failure and named line. Base each hypothesis on an observation and try to refute it before editing. Make independent observations together; change one thing at a time where changes would hide each other's effect. Narrow until one cause explains every symptom; one explaining only some is wrong or one of two. When the error and reported case do not settle narrowing, read narrowing techniques.

Fix the cause

Change what causes the failure, not where it shows: no special case for the reported input, no catch that hides the error, no retry in place of an explanation. Then run the reproduction again, and the tests of the code you changed, since a fix can break a neighbouring case. Keep the reproduction as a test where a test adds signal (read tests-worth-keeping only where its trigger holds). Before you call it fixed, look for the same defect where the same pattern repeats (sibling functions, other callers of the same code) and fix each confirmed instance inside the task's scope; report others without changing them, and say which you fixed.

Failed fix and report

If a fix fails with no new evidence, return to observations and list and check unobserved assumptions; do not vary the same fix. Report reproduction and its command, cause and evidence, refuted hypotheses a later session might retry and their refuting observations, changes and the check passed; say which of reproduction, cause and check nothing you ran shows.

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.