agentleFS
danielgefendanielgefen/sous/skills/file/SKILL.md

file

Skill0 starsChanged 3 months agoMITView on GitHub

File a bug or task from one line. Creates a task in your configured backend with the right labels so the crew picks it up. Usage - /file [P0|P1|P2] <description>

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. mementomori-co/maker-mastery★ 2same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository10 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    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:

    Read sous.config.yaml for the task backend, the repo or project identifier, and the label names.

    line 12
  2. Very specific:

    Read backends/task/<configured backend>.md for the concrete commands.

    line 15
  3. Very specific:

    Run task.search_text with the distinctive nouns from the description, across both open and closed tasks.

    line 26
  4. Distinctive:

    Do not skip this. Duplicate tasks are the most common way an agent queue degrades: two workers pick up the same problem, produce two …

    line 30
  5. Distinctive:

    Do not invent reproduction steps you did not observe.

    line 42
  6. Very specific:

    Call task.create with the title, body, and labels: the configured task label, the ready label, and a priority label if one was given.

    line 47
  7. Distinctive:

    Never create a task without checking for duplicates first.

    line 57
  8. Distinctive:

    Never put secrets or sensitive data in a task.

    line 58
  9. Distinctive:

    If the backend call fails, say so plainly and show the error.

    line 60

≈ “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

File a task

One line is enough. This is the front door to the pipeline.

Step 1 — Load config

Read sous.config.yaml for the task backend, the repo or project identifier, and the label names. If it is missing, tell the user to run /sous:hire first and stop.

Read backends/task/<configured backend>.md for the concrete commands. Never inline backend commands in this skill — that is what makes it work on both GitHub Issues and Jira without being edited.

Step 2 — Parse the request

Extract an optional leading priority token — P0, P1, or P2, case-insensitive. Everything after it is the description. No token means no priority label; do not invent one.

Step 3 — Check for duplicates

Run task.search_text with the distinctive nouns from the description, across both open and closed tasks. If something looks like the same request, show it and ask before creating a second one.

Do not skip this. Duplicate tasks are the most common way an agent queue degrades: two workers pick up the same problem, produce two changes, and one of them wastes a full review cycle.

Step 4 — Write a real title and body

Turn the one line into something an engineer with no context can act on:

  • Title: specific and searchable. "Login button does nothing on Safari", not "login bug".
  • Body: what was asked, what is expected instead, and any context from the current conversation the user did not bother to repeat. If you are mid-session and already know the file or the symptom, include it — that context is free now and expensive to reconstruct later.

Do not invent reproduction steps you did not observe. Write "reported, not reproduced" when that is the truth. A confident fake repro sends the next person down the wrong path.

Step 5 — Create it

Call task.create with the title, body, and labels: the configured task label, the ready label, and a priority label if one was given.

Step 6 — Report

Print the task id, title, and URL, and say the crew will pick it up. Nothing else — this is a 30-second interaction, not a status report.

Rules

  • Never create a task without checking for duplicates first.
  • Never put secrets or sensitive data in a task. Tasks are usually more widely visible than code, and they are rarely audited.
  • If the backend call fails, say so plainly and show the error. Never report a task created that you did not see created.

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.