Skill code-improver
✓ OficialExecuta um loop autônomo de revisão e correção sobre alvos de código usando um agente ou skill de revisão específico até que os requisitos de revisão sejam atendidos.
- Estrelas no GitHub
- 8.136
- Arquivos
- 1
- Último commit
- 11 de nov. de 2026
- Licença
- CC-BY-SA-4.0
- Ferramentas permitidas
- 6
O que a skill code-improver faz?
Executa um loop autônomo de revisão e correção sobre alvos de código usando um agente ou skill de revisão específico até que os requisitos de revisão sejam atendidos.
Resumo gerado a partir do SKILL.md da skill.
Sobre a skill code-improver
Desenvolvedores podem automatizar o processo iterativo de refinamento de código executando um loop autônomo de revisão e correção. Esta skill funciona coordenando um revisor nomeado, que pode ser qualquer agente ou skill instalada, com um subagente de correção para resolver problemas até que uma revisão limpa seja alcançada. Ela foi projetada para melhorar alvos específicos, como diretórios, módulos ou plugins, aplicando correções repetidamente e verificando-as contra um conjunto de requisitos.
O processo depende da ferramenta Workflow para executar um script em segundo plano que gerencia um registro de descobertas, aplica proteções de escopo e lida com a escalação caso as correções não consigam convergir. Ele utiliza ferramentas como Bash Glob, Read e TaskOutput para gerenciar o ambiente. Diferente da edição manual, este workflow garante que todas as alterações permaneçam no working tree sem realizar commits, fornecendo feedback estruturado sobre se o loop convergiu, atingiu um limite máximo de rodadas ou parou devido a violações de escopo.
O que um agente pode fazer com ela
- 1Refinamento iterativo de códigoExecute um loop contínuo para melhorar um diretório específico até que um revisor designado não encontre problemas críticos.
- 2Aplicação automatizada de qualidadeGaranta que o código atenda a padrões específicos através de um loop entre revisor e corretor até que não restem descobertas importantes.
Escrito a partir do SKILL.md da skill. Cada caso de uso é verificado com uma citação dele.
Como instalar a skill code-improver
Cada comando copia a pasta da skill para o diretório de skills do cliente. Reinicie o cliente ou inicie uma nova sessão para carregá-la.Instalar a skill code-improver no Plugin
/plugin marketplace add trailofbits/skills /plugin install code-improver@trailofbits
Instalar a skill code-improver no Claude Code
~/.claude/skills/code-improver/SKILL.md
$ npx degit trailofbits/skills/plugins/code-improver/skills/code-improver ~/.claude/skills/code-improverInstalar a skill code-improver no Claude Code (project)
.claude/skills/code-improver/SKILL.md
$ npx degit trailofbits/skills/plugins/code-improver/skills/code-improver .claude/skills/code-improverInstalar a skill code-improver no Codex
~/.agents/skills/code-improver/SKILL.md
$ npx degit trailofbits/skills/plugins/code-improver/skills/code-improver ~/.agents/skills/code-improverInstalar a skill code-improver no Gemini CLI
~/.gemini/skills/code-improver/SKILL.md
$ npx degit trailofbits/skills/plugins/code-improver/skills/code-improver ~/.gemini/skills/code-improverInstalar a skill code-improver no Claude.ai
npx degit trailofbits/skills/plugins/code-improver/skills/code-improver code-improver && zip -r code-improver.zip code-improver # claude.ai → Customize → Skills → + → Create skill → Upload a skill → code-improver.zip
Verificações automatizadas
Executado em 15 de nov. de 2026Relatório completo →Ainda não verificada. Uma skill precisa de 80 pontos, nenhum achado grave e dados suficientes para a verificação. Como as verificações funcionam · O que corrigir para ganhar mais pontos
- Repositório de origem vinculado e acessíveltrailofbits/skills/plugins/code-improver/skills/code-improver is public and reachablegithub.com/trailofbits/skills/tree/HEAD/plugins/code-improve ↗Passou
- Licença de código aberto (OSI)CC-BY-SA-4.0 is not OSI-approvedspdx.org/licenses/CC-BY-SA-4.0.html ↗Problema encontrado
- Commit ou release no último anoLast commit 2026-11-11github.com/trailofbits/skills/commits ↗Passou
- Sem baixar e executar nem roubo de credenciais nos scriptsNo scripts in the skill folderPassou
- Sem instruções ocultas no SKILL.mdNo override phrases, safeguard switches, credential requests or hidden charactersgithub.com/trailofbits/skills/blob/HEAD/plugins/code-improve ↗Passou
- Ferramentas pré-aprovadas são restritasBash lets the agent run any shell command without askinggithub.com/trailofbits/skills/blob/HEAD/plugins/code-improve ↗Problema encontrado
- O SKILL.md segue o formato Agent SkillsValid name and description; says when to use it; 122 linesgithub.com/trailofbits/skills/blob/HEAD/plugins/code-improve ↗Passou
SKILL.md
Ver no GitHubCode Improver
Improve any code target by running /code-improver:improve — a dynamic workflow that
loops the named reviewer and a fixer subagent until a review reports zero critical/major
findings, then strips its own residue. The loop, its ledger, and its guards live in the
workflow; this skill collects the three inputs the generic entry requires and relays the
outcome.
Starting the loop
The user provided: $ARGUMENTS (if empty, take the details from the conversation).
1. Collect the three required inputs — no guessing
- Target: the absolute path to the directory under improvement. Resolve relative paths against the working directory; verify the directory exists.
- Reviewer: the installed skill or agent that performs every review. The user must
name it — there is no default and no bundled reviewer. Determine the kind:
- a namespaced agent (e.g.
plugin-dev:skill-reviewer) →"kind": "agent" - an installed skill (e.g.
pr-review-toolkit:review-pr) →"kind": "skill"If the name could be either, check the session's skill listing; if still ambiguous, ask the user. If no reviewer was named, ask — do not pick one.
- a namespaced agent (e.g.
- Scope: repo-relative globs the loop may touch. The generic entry requires it
explicitly; if the user did not give one, propose the target directory
(
<repo-relative-target>/**) and confirm before launching.
2. Resolve the loop script
Resolve improve.js from this SKILL.md's installed location
to an absolute path and confirm the file exists. Its plugin directory is pluginRoot.
Use this exact plugin copy; if the file is missing, stop.
3. Invoke the workflow
Run it with the Workflow tool, {scriptPath: "<the path from step 2>", args: {...}}:
{
"target": "<absolute target path>",
"reviewer": { "kind": "agent|skill", "name": "<namespaced-name>", "notes": "<what the reviewer should know about the target>" },
"scope": ["<repo-relative-glob>/**"],
"pluginRoot": "<the plugin directory from step 2>",
"maxRounds": 5
}
maxRoundsonly if the user asked for a different cap.pluginRootis the resolved plugin directory from step 2, for the metrics collector.finalize({"version_bump": bool, "narration_strip": bool, "docs_pass": bool}) only to override the defaults: version bump when the target sits inside a plugin, narration strip and docs pass always.decisiononly on continuation (below).
The workflow runs in the background and needs no babysitting: it reviews, fixes, re-reviews, checks scope after every fix round, and can only complete on a clean review. It never commits; all changes stay in the working tree.
If the Workflow tool is unavailable or denied, stop and say so. Do not improvise the loop inline with direct edits — the ledger, scope guard, and escalation guarantees live in the workflow, and an inline imitation has none of them.
If the result is halted: "reviewer-unavailable", relay it and stop. The named
reviewer is not installed in this session; tell the user which plugin provides it and
re-run after installing. Do not review the target yourself.
Do not end your turn while the loop is running. The Workflow tool returns a task id immediately; the result comes later. In an interactive session the completion notification re-invokes you — wait for it. In a non-interactive run (scripted, CI, eval) there is no later turn: stopping abandons the loop mid-round, so after launching, poll the task (TaskOutput with the returned task id, or sleep-and-recheck) until it completes, then relay the result. A session that answers "the loop is running, I'll report later" has lost the run.
Relaying the result
The workflow returns a structured result. Report it honestly — the distinctions matter:
converged: true— the last action was a review with zero critical/major findings. Report rounds used, remaining minor findings (open_minor_count), and the artifact paths (ledger_path,metrics).capped: true— the fix budget ran out and the FINAL review still found blocking issues. Say plainly: capped, NOT converged, and listopen_blocking. Do not present this as success.escalation— the loop detected it was not converging (recurring findings, non-decreasing counts, or a fix relocating a problem). Relay the escalation message and finding ids to the user: this needs a design decision, not more rounds.halted— a guard fired (scope violation, unregistered new files, a dead or unavailable reviewer, or a finalize pass whose own edits failed the check that follows it). Relay the paths inviolations/new_untracked_files, the sites infinalize_regressions, and the notes.notesalways travel with the result — surface them; they include loud warnings such as "a git repository was initialized".
Continuing after an escalation
The loop stops on escalation by design. When the user decides, start a fresh run with
the same target and reviewer plus:
{ "decision": "<the user's ruling, verbatim>" }
The new run reloads the on-disk ledger, so every finding, rejection, and verdict carries over — rounds restart, re-derivation does not.
To stop a running loop, stop the workflow task (TaskStop); the ledger on disk is current to the last round and a re-run resumes from it.
When NOT to use
- A Claude Code skill: use the
skill-improverentry — it wires the right reviewer - A branch / pull request: use the
pr-improverentry — it derives scope from the diff - One-time review: dispatch the reviewer directly; the loop's value is iteration
- Quick single fixes: edit the file directly
Arquivos desta skill
1Perguntas sobre a skill code-improver
Como instalo a skill code-improver no Claude Code?
Execute npx degit trailofbits/skills/plugins/code-improver/skills/code-improver ~/.claude/skills/code-improver. O Claude Code carrega skills de ~/.claude/skills (todos os projetos) ou .claude/skills (um projeto) e usa uma skill quando a descrição dela corresponde à tarefa.
Quais clientes podem usar a skill code-improver?
Qualquer cliente que leia o formato Agent Skills: Claude Code, Codex, Gemini CLI e claude.ai (envie como ZIP). Scripts da skill são executados apenas onde o cliente permite execução de código.
O que acontece se o loop de melhoria parar de progredir?
O sistema aciona uma escalação se detectar que o loop não está convergindo, como descobertas recorrentes ou contagem de problemas sem diminuição, exigindo uma decisão de design do usuário.
A ferramenta faz commit automático das alterações no meu repositório?
Não, o workflow nunca faz commit; todas as alterações feitas durante o loop de melhoria permanecem no working tree.