All skills
simota avatar

/scribe

@35ffd55
by shingo imotasimota/agent-skills85 stars
15

Authoring standalone and cross-team specifications: PRD/SRS/HLD/LLD, staged L0-L4 unified packages, BDD acceptance criteria, and traceability. Use for technical or multi-audience documentation; not implementation or architecture decisions.

Use this Skill: https://skilld.dev/gh/simota/agent-skills/scribe

This session only. Nothing lands on disk.

referenceunified-specuser-story-smells.md

≈646 tokens on demand. Your agent reads this file only when SKILL.md points to it.

User Story Smells And Prioritization Pitfalls

Purpose: Use this file when stories, backlog slices, or MoSCoW priorities look weak or inflated.

Contents

  • USS-01..09
  • MP-01..05
  • Backlog anti-patterns
  • Scribe quality gates

User Story Smells

ID Smell Failure mode Guardrail
USS-01 everything must be a story technical work is forced into the wrong format allow the right work item format
USS-02 multi-page story story becomes a mini requirements document keep the story as a conversation anchor
USS-03 missing acceptance criteria completion is undefined add AC before work starts
USS-04 AC/DoD confusion feature and team-wide quality are mixed keep AC feature-specific
USS-05 giant story cannot finish within one sprint split vertically
USS-06 technical-layer split no end-to-end value slice by user outcome
USS-07 no ready definition under-refined work enters delivery require value, size, AC, and context
USS-08 no refinement details freeze too early or never mature refine continuously
USS-09 forgotten conversation story becomes a static document keep the conversation active

MoSCoW Pitfalls

ID Pitfall Failure mode Guardrail
MP-01 everything is Must prioritization collapses keep Must <= 60%
MP-02 stakeholder disagreement teams rank differently align priorities together
MP-03 one-time prioritization priorities do not respond to change re-evaluate iteratively
MP-04 urgency equals importance strategic work is displaced separate urgency from importance
MP-05 no success definition ranking is subjective derive priority from L0 KPI and value

Must Ratio Check

Warning signs:

  • Must exceeds 60%
  • Could or Won't are nearly empty
  • teams say everything blocks release

Backlog Anti-Patterns

Pattern Failure mode Scribe response
bloated backlog 100+ items and no usable priority order trigger decomposition
stale backlog old items remain forever use Deprecated status
estimate everything up front high change cost estimate progressively
no epic or theme layer no visible business linkage anchor in L0 and L1
hidden tech debt risk stays invisible expose it in L2-Dev

Scribe Quality Gates

  • no AC -> not ready
  • Must > 60% -> warning
  • technical-layer slicing -> recommend vertical slicing
  • priority with no KPI rationale -> warning

Source: SKILL.md on GitHub

1 warning13d5 checks · Risk SAFE
  • Gen Agent Trust Hub13d

    No security issues were detected. The skill is designed for authoring technical specifications and documentation. It integrates standard format conversion tools and follows industry best practices for requirements engineering and AI-agent compatibility.

  • Socket13d

    No alerts

  • Snyk13d

    Risk: LOW · No issues

  • Runlayer6mo

    1/8 files flagged

  • ZeroLeaks5mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at 35ffd55. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 days ago.

Activeupdated 2 weeks ago

README badge

README badge for simota/agent-skills/scribe