agentsclimarketplace

Documentation

Skill linenoize/topia/skills/documentation

Leadership-ready package generator. Produces feature folders, requirements, user stories, and Jira-compliant CSV tickets. Use after /topia plan and before /topia build to get stakeholder approval.From its SKILL.md

Install
npx -y skills add linenoize/topia --skill documentation

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

5.7 KB, as published. Nobody here has run it

documentation

Purpose

Bridges the gap between technical planning and organizational approval. This skill synthesizes the output of idea (Requirements) and plan (Blueprints) into a professional "Leadership Package". It ensures that stakeholders have a clear, non-technical view of the work while providing engineers with importable Jira tickets.

Triggers

  • User explicitly runs /topia documentation
  • Automatically suggested after plan or adversary
  • User asks for "Jira tickets", "CSV", or "stories"

Calls (outbound)

  • idea (L2): read requirements.md from feature folder
  • plan (L2): read plan.md and phase files from feature folder
  • adversary (L2): read adversary-report.md when available for risk summary

Called By (inbound)

  • build (L1): when leadership package needed before implementation
  • plan (L2): when plan output needs stakeholder-ready artifacts
  • User: /topia documentation direct invocation

Workflow

Phase 1: Context Intake

  1. Identify Target: Locate the active feature folder in .topia/features/<name>/.
  2. Read Inputs:
    • Load requirements.md (from idea).
    • Load plan.md and phase files (from plan).
    • Load adversary-report.md (if available).

Phase 2: User Story Generation

  1. Draft Stories: Convert phase tasks into Agile User Stories.
  2. Format: As a <role>, I want to <action>, so that <value>.
  3. Criteria: Map each phase's Acceptance Criteria to its corresponding story.

Phase 3: Jira CSV Generation

  1. Adhere to Standards: Follow docs/JIRA_CSV_INSTRUCTIONS.md strictly.
  2. Map Fields:
    • Issue Type: Map phases to Feature (user-facing) or Task (internal).
    • Summary: 10-15 word sentence case summary.
    • Description: Must include both non-technical and technical layers.
    • Priority: Set based on impact (default: Medium).
    • Labels: Add relevant domain/surface labels.
  3. Write File: Save to .topia/features/<name>/jira-tickets.csv.

Phase 4: Leadership Briefing

  1. Generate Brief: Create leadership-brief.md in the feature folder.
  2. Include:
    • Executive Summary: 2-sentence value proposition.
    • Timeline: Estimated phases.
    • Risk Assessment: Summary of findings from adversary.
    • Proof of Quality: Mention planned test coverage.

Constraints

  1. Strict CSV Format: The Jira CSV must NOT have markdown around it when emitted as a file.
  2. Stakeholder Language: Descriptions in the "Overview" section must be devoid of jargon.
  3. No Code: This skill generates documentation and metadata only; it does not implement features.
  4. Consistency: Summary in CSV must match the intent of the corresponding phase in the plan.

Done When

  • Feature folder .topia/features/<name>/ contains:
    • user-stories.md
    • jira-tickets.csv (RFC-4180 compliant)
    • leadership-brief.md
  • Total ticket count matches the number of implementation phases.

Sharp Edges

Known failure modes for this skill. Check these before declaring done.

Failure ModeSeverityMitigation
Generating tickets without requirements.md and plan.md presentCRITICALPhase 1 MUST locate both inputs; if missing, abort and tell user to run topia:idea + topia:plan first
Jira CSV not RFC-4180 compliant (unescaped commas, quotes, newlines inside fields)CRITICALQuote every field; escape embedded " as ""; verify with a CSV linter before declaring done
Markdown fences around the CSV file contentHIGHConstraint 1 — the .csv file must be raw CSV, not wrapped in ```; only the preview shown to the user may be fenced
Phase count ≠ ticket countHIGHDone-When requires 1:1 mapping; missing or duplicate tickets break import
User stories that paraphrase code instead of stating user valueHIGHStory format As a <role>, I want <action>, so that <value> — the value clause is mandatory and non-technical
Leadership brief contains technical jargonMEDIUMConstraint 2 — Overview/Executive Summary in stakeholder language; technical detail belongs in ticket Description, not the brief
Acceptance criteria copy-pasted from plan instead of restated for testersMEDIUMEach story's AC must be testable by a non-author; rewrite plan's HOW into GIVEN/WHEN/THEN
Skipping adversary-report.md when it existsMEDIUMRisk Assessment section in the brief loses signal if adversary findings are not summarized
Writing implementation code to satisfy a Jira fieldHIGHConstraint 3 — documentation only. If a field needs code-level detail, point to a file path, don't inline code
Stale tickets after plan revisionMEDIUMIf plan.md is regenerated, re-run documentation to keep ticket/phase parity

Output Format

LEADERSHIP PACKAGE — <feature-name>
===================================
Feature folder: .topia/features/<name>/

Artifacts:
- user-stories.md — Agile stories with acceptance criteria
- jira-tickets.csv — RFC-4180 import file (N tickets)
- leadership-brief.md — Executive summary, timeline, risks

Ticket summary:
| Phase | Issue Type | Summary |
| ...   | ...        | ...     |

Cost Profile

~1000-2000 tokens input, ~1000 tokens output. Sonnet for professional writing quality.

Keep looking

Skills are one crate of 627,801. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.