ag-melhorar-agentes

ag-melhorar-agentes is a skill for Claude Code from andregusman-raiz/a-gusman-claude. It costs 36 tokens per session (2,191 once invoked), scanned A, original, MIT.

A review process for improving other coding agents by studying their real reports, error logs, and evaluation results. It looks for repeated failures and connects proposed prompt changes to evidence.

In plain words
What is it for?
Use it to diagnose a specific agent, review the whole agent system, run benchmarks, or improve an agent's description so it triggers in the right situations.
Why use it?
It helps prevent agent instructions from being changed based only on assumptions. The goal is to find recurring problems and make improvements that can be measured.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: model in frontmatter; reads .claude/ paths.

Good fit Use it to diagnose a specific agent, review the whole agent system, run benchmarks, or improve an agent's description so it triggers in the right situations.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/andregusman-raiz/a-gusman-claude/ag-melhorar-agentes
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add andregusman-raiz/a-gusman-claude --skill ag-melhorar-agentes
Clone the repo
git clone --depth 1 https://github.com/andregusman-raiz/a-gusman-claude

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ag-melhorar-agentes

README.md
[![agentmods](https://agentmods.dev/badge/skills/andregusman-raiz/a-gusman-claude/ag-melhorar-agentes/github.svg)](https://agentmods.dev/skills/andregusman-raiz/a-gusman-claude/ag-melhorar-agentes)
Your own site
<a href="https://agentmods.dev/skills/andregusman-raiz/a-gusman-claude/ag-melhorar-agentes"><img src="https://agentmods.dev/badge/skills/andregusman-raiz/a-gusman-claude/ag-melhorar-agentes/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ag-melhorar-agentes

Your own site · 80×15
<a href="https://agentmods.dev/skills/andregusman-raiz/a-gusman-claude/ag-melhorar-agentes"><img src="https://agentmods.dev/badge/skills/andregusman-raiz/a-gusman-claude/ag-melhorar-agentes.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 36 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,191 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium Excessive Agency · line 4
    Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.
    Fix: Remove the model/provider override or disclose it prominently and require explicit operator approval before invoking an external coding CLI or billed model.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00036 $0.02191
Opus 5 $0.00018 $0.01095
Sonnet 5 $0.00007 $0.00438
Haiku 4.5 $0.00004 $0.00219

Measured 9d ago against content hash ddcd090b7b8d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

ag-melhorar-agentes scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/ag-melhorar-agentes/SKILL.md · 157 lines

How it starts

The opening of the file, as written. The whole thing — 157 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ag-melhorar-agentes — Melhorar Agentes

Reasoning protocol (tier topo — Fable, equivalente a reasoning_effort=xhigh): Mudar prompts de agentes sem evidencia = regressao. Exhaust 3+ hipoteses para cada padrao de falha; Verify com leitura de session logs / reports reais; Falsify (a falha seria explicada por outra causa que nao o prompt?); Connect cada mudanca proposta a uma falha documentada; Report melhorias com baseline mensuravel. Mudanca sem evidencia em logs = bloqueio. Detalhes: .claude/rules/deep-reasoning-directive.md.

Quem você é

O Meta-Agente. Analisa como os outros agentes trabalham e melhora seus prompts.

Modos

/ag-melhorar-agentes diagnosticar [ag-XX] → Analisar reports de um agente específico
/ag-melhorar-agentes calibrar → Avaliar todos os agentes
/ag-melhorar-agentes panorama → Visão geral do sistema
/ag-melhorar-agentes benchmark [ag-XX] → Rodar evals quantitativos via ag-criar-skill
/ag-melhorar-agentes otimizar-description [ag-XX] → Otimizar triggering via ag-criar-skill

Fontes de dados

  1. docs/ai-state/errors-log.md → Padrões de falha recorrentes
  2. validation-report.md → O que o ag-validar-execucao encontra repetidamente
  3. e2e-report.md → Bugs que escapam para o E2E
  4. test-report.md → Cobertura e falhas

Princípios

  • Melhora o PROMPT que produz o comportamento, não o comportamento em si
  • Prefere explicar "porquê" a adicionar regras
  • Nunca sacrifica generalidade por caso específico
  • Comparação cega (blind A/B) entre versões de prompt

Output

Proposta de melhoria com: evidência, rationale, risco documentado.

Integracao com ag-criar-skill

Para avaliacao quantitativa de skills, ag-melhorar-agentes delega ao ag-criar-skill:

  • benchmark [ag-XX]: Cria test cases, roda evals com/sem skill, gera grading + benchmark
  • otimizar-description [ag-XX]: Roda loop automatizado de calibracao de triggering
  • Grader agent: ~/.claude/skills/ag-criar-skill/agents/grader.md — avalia assertions
  • Comparator agent: ~/.claude/skills/ag-criar-skill/agents/comparator.md — blind A/B
  • Analyzer agent: ~/.claude/skills/ag-criar-skill/agents/analyzer.md — post-hoc analysis
  • Viewer: ~/.claude/skills/ag-criar-skill/eval-viewer/generate_review.py — HTML review
  • Workspace: ~/.claude/skills-workspace/[skill-name]/ — resultados de evals

Read the full file on GitHub · 157 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 157 lines · 36 tokens per session scan A ddcd090b7b8d

Subscribe to this mod's changes

ag-melhorar-agentes is a skill published in the GitHub repository andregusman-raiz/a-gusman-claude (19 stars, last pushed 4d ago), licensed MIT. It adds 36 tokens to every session and 2,191 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Related

Other skills, from other repositories

taiyi-ui-design

A design-planning guide for describing how an application's user interface should look and behave. It produces a UI-DESIGN.md document covering layouts, components, interactions, accessibility, and error states.

Dong90/oh-my-taiyiforge · 35 tokens

taiyi-evolve

A workflow skill that compares the implemented code with the frozen design after development and testing. It records architecture changes and proposes updates to DESIGN.md, the document describing the intended system structure.

Dong90/oh-my-taiyiforge · 37 tokens

taiyi-diagram-c4

A code-scanning tool that builds C4 architecture documents from a repository. It separates facts observed in the code from conclusions inferred about the design and uses Mermaid diagrams as the source format.

Dong90/oh-my-taiyiforge · 48 tokens

retrospective-audit

Stage B of /prflow:retrospective-weekly: given a most-recent-first subset of one recurring pattern's occurrence-PR context bundles (bounded by auditbundlecap), re-derive the root cause and return one JSON object carrying a ranked findings array (one to three sub-patterns) — no edits, no worktree. Invoked as a subagent…

The01Geek/prflow · 91 tokens

docs

Use when documentation generally needs to catch up with a branch before pushing or merging, covering internal developer docs, external customer-facing docs, and release notes together — "update the docs", "do a docs pass before I merge", "make sure everything's documented". Prefer this when no single documentation…

The01Geek/prflow · 97 tokens

tmux-lane-orchestrator

Manage one tmux agent lane from its matching ops pane, inspect live pane state and Codex logs on cold start, and produce concise manager summaries for OpenClaw and adjacent project work.

vincentkoc/dotskills · 46 tokens