agentleFS
jaroslawherodjaroslawherod/micold-ai-ide/.claude/skills/speckit-bugfix-patch/SKILL.md

speckit-bugfix-patch

Skill1 starsChanged 3 days agoApache-2.0View on GitHub

Surgically update spec, plan, and tasks to address the reported bug

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 8 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    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:

    Verify a spec-kit project exists by checking for .specify/ directory

    line 28
  2. Very specific:

    Check for bug reports in specs/{feature}/bugs/ — if a bug ID is provided, load that report

    line 30
  3. Very specific:

    If no bug report exists, inform the user and suggest running /speckit.bugfix.report first

    line 31
  4. Distinctive:

    Load bug context: Read the relevant bug report and all spec artifacts:

    line 35
  5. Distinctive:

    Add missing requirements under the affected user story

    line 51
  6. Distinctive:

    Add success criteria for untested flows

    line 53
  7. Distinctive:

    Update assumptions if dependencies changed

    line 54
  8. Distinctive:

    Add a bugfix note:

    line 55
  9. Distinctive:

    Update affected sections with new context

    line 61
  10. Distinctive:

    Add complexity notes for newly discovered edge cases

    line 62
  11. Very specific:

    Mark false completions: Add ⚠️ Reopened prefix to task description

    line 72
  12. Distinctive:

    Update Wave DAG: If present, regenerate to include new tasks

    line 73

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Bugfix Patch Skill

Patch Spec Artifacts

Surgically update spec.md, plan.md, and tasks.md to address a reported bug — adds missing requirements, fixes conflicts, reopens false completions, and adds new tasks. Minimal changes only, never regenerates from scratch.

User Input

$ARGUMENTS

You MUST consider the user input before proceeding (if not empty). The user may specify a bug report to patch (e.g., "BUG-001") or describe the fix directly.

Prerequisites

  1. Verify a spec-kit project exists by checking for .specify/ directory
  2. Locate the current feature's spec directory
  3. Check for bug reports in specs/{feature}/bugs/ — if a bug ID is provided, load that report
  4. If no bug report exists, inform the user and suggest running /speckit.bugfix.report first

Outline

  1. Load bug context: Read the relevant bug report and all spec artifacts:
    • Bug report: specs/{feature}/bugs/BUG-{NNN}.md (if specified)
    • Required: spec.md, and at least one of plan.md or tasks.md
    • Optional: research.md, data-model.md
  2. Determine patches: Based on the bug type, plan minimal changes:

    Bug Typespec.md Patchplan.md Patchtasks.md Patch
    Spec gapAdd missing requirement to affected user storyAdd implementation note to relevant sectionAdd new task(s) for the missing requirement
    Spec conflictResolve conflict with strikethrough on superseded text + new clarified requirementUpdate affected sectionUpdate affected task descriptions
    Implementation driftAdd clarification note to requirementNo change (plan was correct)Reopen drifted task with correction note
    Untested flowAdd success criterion for the edge caseAdd edge case to complexity trackingAdd verification task
    Dependency issueUpdate assumption about external dependencyUpdate technical contextAdd dependency investigation task
  3. Patch spec.md:
    • Add missing requirements under the affected user story
    • Mark conflicting text with ~~strikethrough~~ and reason
    • Add success criteria for untested flows
    • Update assumptions if dependencies changed
    • Add a bugfix note:
      **Bugfix**: [DATE] — [BUG-NNN] [Brief description of what was patched]
  4. Patch plan.md (if it exists):
    • Update affected sections with new context
    • Add complexity notes for newly discovered edge cases
    • Preserve all existing content — only add or annotate
    • Add a bugfix note:
      **Bugfix**: [DATE] — [BUG-NNN] Updated from bugfix patch
  5. Patch tasks.md (if it exists):
    • Add new tasks: Assign next sequential IDs, proper dependencies, and story labels
    • Reopen tasks: Change [x] back to [ ] with a note: (reopened — BUG-NNN)
    • Mark false completions: Add ⚠️ Reopened prefix to task description
    • Update Wave DAG: If present, regenerate to include new tasks
    • Add a bugfix note:
      **Bugfix**: [DATE] — [BUG-NNN] Updated from bugfix patch
  6. Update bug report: Mark the bug report file as patched:
    **Status**: Patched
    **Patched**: [DATE]
  7. Report: Output a summary:
    • What changed in each artifact
    • How many requirements were added or updated
    • How many tasks were added or reopened
    • Suggest next step: /speckit.bugfix.verify to confirm consistency, then /speckit.implement to apply the code fix

Rules

  • Surgical updates only — never regenerate artifacts from scratch, only modify affected sections
  • Never delete content — use strikethrough for superseded text, preserve history
  • Preserve formatting — match existing artifact style exactly
  • Track changes — always add bugfix notes with dates and bug IDs
  • Reopen, don't delete tasks — falsely completed tasks get reopened, not removed
  • Require bug report — if no bug report or user description is provided, refuse to patch and suggest /speckit.bugfix.report first
  • Minimal changes — change only what is necessary to address the specific bug

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.