agentleFS
Sign inStart free
duviolinduviolin/sdd-kit/skills/sdd-report/SKILL.md

sdd-report

Skill0 starsChanged 54 days agoMITView on GitHub

Use to generate a usage & metrics report of the SDD work on a project or feature. Auto-use when the user says "relatório", "report", "métricas", "quanto custou", "quanto gastei", "tokens gastos", "tempo gasto", "ROI do SDD", or wants to justify/analyze SDD usage after delivery. Reads local transcripts + SDD artifacts (nothing leaves the machine) and writes report.md.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. pchemguy/Skill-SDD-Manager★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository9 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ler os números, não só listar: No resumo executivo, dizer o que a métrica significa (ex.: "78% do gasto foi em decisão → o erro foi …

    line 46

≈ “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

sdd-report

Gera um relatório de esforço (tokens, tempo, custo) + qualidade/governança do uso do SDD, cruzando os transcripts locais do Claude Code com os artefatos SDD do repo. Emite report.md.

Tudo é local: lê só ~/.claude/projects/... e arquivos do projeto; não envia nada.

Escopo

  • Feature: relatório de uma feature — usa a branch da feature (feat|fix|chore/<slug>) e specs/<NNN>-<slug>/. Grava specs/<NNN>-<slug>/report.md.
  • Projeto: relatório consolidado de todas as branches/features. Grava report.md na raiz.
  • Na dúvida, perguntar qual escopo; se houver uma feature ativa no state.md, assumir ela.

Passos

  1. Esforço (determinístico). Rodar o agregador e capturar o JSON: python3 ${CLAUDE_PLUGIN_ROOT}/sdd/report/usage.py --project-dir <raiz-do-projeto> [--branch <branch-da-feature>] Ele devolve, por feature/modelo/fase: turnos, tool calls, tokens (in/out/thinking/cache), tempo (ativo e decorrido), web calls e custo estimado (via sdd/report/pricing.json).
    • Sem Python 3, ou transcript_files: 0 / warning no JSON: dizer isso no relatório e seguir só com a parte de qualidade (não inventar números).
  2. Qualidade & governança (dos artefatos). Ler, quando existirem:
    • tasks.md → nº de tasks, quantas done com evidência objetiva, re-trabalho (reabertas/blocked).
    • state.md → histórico: checkpoints humanos, gates do sdd-analyze, o que foi retomado.
    • requirements.md → checkpoint de aprovação do pedido.
    • design.md → guidelines aplicadas.
    • acceptance.md → veredito final e conformidade com a constitution.md.
  3. Compor. Preencher ${CLAUDE_PLUGIN_ROOT}/sdd/templates/report.md com os dois lados e gravar o report.md no caminho do escopo. Derivar:
    • hh:mm a partir dos segundos; R$ a partir de US$ × fx_to_brl (se fx_to_brl != null).
    • custo por task = custo / nº de tasks entregues.
    • % pensar vs executar = custo das fases de decisão (constitution, discovery, requirements, design, acceptance) sobre o total, vs execução (backlog, analyze, delivery) + registro (state).
  4. Ler os números, não só listar. No resumo executivo, dizer o que a métrica significa (ex.: "78% do gasto foi em decisão → o erro foi decidido no texto, barato").

Regras

  • Só reporte o que sai do agregador ou dos artefatos. Nada de estimar tokens/tempo de cabeça.
  • Custo é sempre ESTIMATIVA — citar sdd/report/pricing.json e a data (as_of); se o preço/plano mudou, o usuário edita esse arquivo. Lembrar que cache-read é barato e domina os tokens, então tokens brutos ≠ custo.
  • Se um lado faltar (sem transcript, ou sem artefatos SDD), gerar o relatório com o que há e dizer o que faltou — não bloquear.
  • O report.md é um registro auditável; deve se sustentar sozinho, sem depender da conversa.
  • Não altera artefatos de outras skills; só escreve report.md. Não precisa de lock nem de branch dona.

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.

Check your setup, free