agentleFS
slashnote-appslashnote-app/claude-code-plugin/skills/note-bug/SKILL.md

note-bug

Skill1 starsChanged 9 months agoMITView on GitHub

Log a structured bug report with git context and recent diff

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    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:

    Create a structured bug report in a SlashNote sticky note with git context.

    line 12
  2. Distinctive:

    Create a pink note with structured format

    line 27
  3. Distinctive:

    If not in a git repo → skip "Branch" and "Recent changes" lines entirely

    line 68
  4. Distinctive:

    If user provides enough detail → fill in Repro steps (infer logical sequence)

    line 80
  5. Distinctive:

    If minimal input → use placeholder checkboxes ("Describe step 1", "Describe step 2")

    line 81
  6. Distinctive:

    Keep it compact — bug notes should be scannable in 5 seconds

    line 82
  7. Distinctive:

    If not in a git repo (Bash returns "no-git"):

    line 133
  8. Distinctive:

    Do NOT ask followup questions — create the note immediately

    line 140
  9. Distinctive:

    Keep the note scannable — developers glance at bugs, they don't read essays

    line 142

≈ “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

/note-bug — Bug Report

Create a structured bug report in a SlashNote sticky note with git context.

Usage

/note-bug <description>

Behavior

  1. Parse the user's bug description
  2. Collect git context with a single Bash call:
    echo "BRANCH:$(git rev-parse --abbrev-ref HEAD 2>/dev/null || echo 'no-git')" && echo "---DIFF---" && git diff --stat HEAD~1 2>/dev/null | tail -5
  3. Create a pink note with structured format

Output Format

# Bug: <concise title (max 8 words)>
**Severity:** <Critical / High / Medium / Low>

## What happens
<1-2 sentence description from user input>

## Context
Branch: <current git branch>
Date: <today's date>
Recent changes: <summary of git diff --stat, e.g. "3 files changed (auth.swift, api.ts, test.ts)">

## Repro
- [ ] <step 1>
- [ ] <step 2>
- [ ] <step 3>

## Expected → Actual
**Expected:** <inferred or "TBD">
**Actual:** <from description>

Severity Inference

Infer severity from the description — do not ask the user:

Keyword SignalsSeverity
"crash", "data loss", "security", "down", "blocks"Critical
"broken", "fails", "can't", "regression"High
"wrong", "incorrect", "unexpected", "slow"Medium
"minor", "cosmetic", "typo", "alignment"Low
No clear signalMedium (safe default)

Git Context Rules

  • One Bash call — combine branch + diff in a single command
  • If not in a git repo → skip "Branch" and "Recent changes" lines entirely
  • Parse git diff --stat output: extract changed file names, summarize as "N files changed (file1, file2, ...)"
  • Show max 3 file names in the summary, use "+N more" for the rest
  • This context helps identify which recent changes may have caused the bug

Formatting Rules

  • Always pink color
  • Title: max 8 words, actionable ("Login crashes on Safari" not "Bug with login")
  • "What happens" instead of "Summary" — more natural
  • "Repro" instead of "Steps to Reproduce" — shorter
  • "Expected → Actual" — compact format
  • If user provides enough detail → fill in Repro steps (infer logical sequence)
  • If minimal input → use placeholder checkboxes ("Describe step 1", "Describe step 2")
  • Keep it compact — bug notes should be scannable in 5 seconds

Examples

Input: /note-bug Login crashes on Safari when clicking submit Result:

# Bug: Login crashes on Safari submit

## What happens
Login page crashes on Safari when clicking the submit button.

## Context
Branch: feature/auth-flow
Date: Feb 19, 2026
Recent changes: 3 files changed (AuthController.swift, LoginView.swift, +1 more)

## Repro
- [ ] Open login page in Safari
- [ ] Enter credentials
- [ ] Click submit button

## Expected → Actual
**Expected:** Successful login redirect
**Actual:** Page crashes on submit

Input: /note-bug race condition in websocket Result:

# Bug: Race condition in WebSocket handler

## What happens
Race condition occurring in WebSocket handler.

## Context
Branch: main
Date: Feb 19, 2026
Recent changes: 2 files changed (WebSocketManager.swift, ConnectionPool.swift)

## Repro
- [ ] Describe trigger condition
- [ ] Describe sequence of events

## Expected → Actual
**Expected:** TBD
**Actual:** Race condition in WebSocket handler

Non-Git Mode

If not in a git repo (Bash returns "no-git"):

  • Skip "Branch" and "Recent changes" lines
  • Context section becomes just Date: <today>
  • Everything else stays the same

Important

  • Do NOT ask followup questions — create the note immediately
  • One Bash call max for git context
  • Keep the note scannable — developers glance at bugs, they don't read essays
  • Severity is inferred, never asked — it's a visual hint, not a SLA commitment

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.