agentleFS
joris887joris887/exosuit/.claude/skills/backlog-review/SKILL.md

backlog-review

Skill9 starsChanged 3 months agoMITView on GitHub

Execute comprehensive backlog review. Analyzes story quality, Definition of Ready compliance, dependencies, zombie stories, and generates a backlog health report.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create dependency graph for upcoming work:

    line 71
  2. Very specific:

    Check each story's created: date (from inline metadata or YAML frontmatter):

    line 88
  3. Very specific:

    If docs/sessions/.activity-log.jsonl exists and contains story events:

    line 98
  4. Very specific:

    Update docs/reference/BACKLOG_INDEX.md Backlog Health section with the calculated metrics.

    line 152

≈ “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

backlog-review

Execute comprehensive backlog review:

1. Load Current State

Read these files:

  • @docs/reference/BACKLOG_INDEX.md (backlog overview)
  • docs/reference/backlog/E##-name.md (epic files as needed)
  • @docs/reference/PRD_SUMMARY.md (requirements overview, if exists)
  • @docs/progress.md (completed work)

2. Theme Coverage Analysis

Map each epic to requirements:

| Epic | Requirements Covered | Coverage % | Gap Analysis |

Priority Validation

  • Are high-priority items truly high-value?
  • Are dependencies properly ordered?
  • Any blocked items that should be deprioritized?

Scope Assessment

  • Stories that exceed original scope
  • Features creeping in without justification
  • Non-goals being violated

3. Story Quality Scan

For each non-done story, validate against the Definition of Ready checklist:

CriterionCheck
Title clear and specificDescribes the change, not the problem
Type assignedOne of: feature, bugfix, refactor, spike, infra, testing, docs, security, performance, skill
Size classifiedTRIVIAL, SMALL, STANDARD, LARGE, or XL. Flag for splitting only when the story spans unrelated topics — file count is not a threshold. See the cohesion test in ideate/references/story-template.md
Cohesion test passedSplitting the story would create artificial seams, not two meaningful pieces
3-7 acceptance criteriaEach testable and specific; LARGE/XL may exceed 3-7, grouped under sub-headings. No "should be fast" or "handle errors properly"
Verification commandsExact commands that prove completion
Out of scope definedAt least one explicit exclusion
Affected files listedProportionate to size, not capped at a fixed number
Pattern references"Follow patterns in [file]" where applicable
Dependencies resolvedNo unresolved blockers
No ambiguous languageNo vague terms, all criteria measurable
Self-containedAll referenced context included or linked

Classify each story:

  • Ready: Passes all 12 criteria → status: ready
  • Needs refinement: Missing 1-3 criteria → list what's missing
  • Needs rewrite: Missing 4+ criteria or too large → flag for /ideate refinement

4. Dependency Mapping

Create dependency graph for upcoming work:

[Story A] --depends on--> [Story B]
                              |
                              v
                         [Story C]

Validate dependency health:

  • Flag circular dependencies (A→B→C→A)
  • Flag stories blocked by draft/not-ready dependencies
  • Flag stories with >2 transitive dependencies (fragile chain)
  • Identify the critical path: longest dependency chain

5. Zombie Story Detection

Check each story's created: date (from inline metadata or YAML frontmatter):

  • Zombie (>2 sprint cycles old, still not done): Must be split or killed
  • Aging (>1 sprint cycle, same status): Flag for attention
  • Stalled (status: in-progress but no commits referencing it in >1 week): Flag as potentially abandoned

For each zombie, recommend: Split (too large), Kill (no longer needed), or Unblock (has a blocker that can be resolved).

6. Flow Metrics

If docs/sessions/.activity-log.jsonl exists and contains story events:

  • Throughput: Stories completed per sprint (from story events with to: done)
  • Cycle time by size: Average time from in-progress to done for TRIVIAL/SMALL/STANDARD/LARGE/XL
  • Completion rate: Stories started vs completed
  • Type distribution: Which story types are most common

7. Backlog Health Report

## Backlog Health Report - [Date]

### Overall Health: X/10

### Definition of Ready Compliance
- Stories meeting DoR: X/Y (Z%)
- Common DoR failures: [list most frequent missing criteria]
- Stories needing refinement: [list IDs]

### Coverage
- Requirements covered: X%
- Orphan stories (no requirement trace): X
- Requirement gaps (no stories): [list]

### Quality
- Stories ready for sprint: X
- Stories needing refinement: Y
- Stories too large (need split): Z

### Flow
- Throughput: X stories/sprint (last 3 sprints)
- Avg cycle time: TRIVIAL Xh, SMALL Xh, STANDARD Xh
- Completion rate: X%

### Zombie Stories
- Zombie (>2 cycles): [list IDs with age]
- Aging (>1 cycle): [list IDs]
- Stalled in-progress: [list IDs]

### Dependency Health
- Stories with unresolved blockers: [list]
- Circular dependencies: [list or "None"]
- Critical path length: X stories

### Priority Issues
- Blocked items: [list]
- Misordered dependencies: [list]

### Recommended Actions
1. [Action with specific story IDs]
2. [Action]
3. [Action]

Update docs/reference/BACKLOG_INDEX.md Backlog Health section with the calculated metrics.

Output the full backlog health report with specific recommendations.

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.