Skip to content

majiayu000/spellbook / skills/agentsmd-scaffold

agentsmd-scaffold

Verified

Generate or update repository-specific AGENTS.md instruction files from real repo evidence. Use when asked to create, design, scaffold, split, or improve root or scoped AGENTS.md files for Codex/Claude/agent workflows, especially when a repo needs directory-specific rules, validation commands, generated-file boundaries, or a short agent onboarding router.

FileKindSize
SKILL.mdskill md5.9 KB
evals/evals.jsonconfig1.4 KB
references/scaffold-agents.mdmarkdown5.5 KB
scripts/scan_repo_context.pyscript13.0 KB

evals/evals.json

View on GitHub
{
  "skill_name": "agentsmd-scaffold",
  "evals": [
    {
      "id": 1,
      "prompt": "I'm in a repo that has README.md, package.json, scripts/, tests/, and generated OpenAPI clients under src/generated. Design the AGENTS.md setup you would add. Don't write files yet; give me the scoped plan and validation commands.",
      "expected_output": "Produces a scoped AGENTS plan with root, scripts, tests, and generated-client candidates; cites evidence to inspect; avoids writing files without approval.",
      "files": []
    },
    {
      "id": 2,
      "prompt": "This Rust workspace has a giant CLAUDE.md at the root and crates/api, crates/cli, migrations, and .github/workflows. Help me split the agent instructions into something Codex can actually follow.",
      "expected_output": "Recommends a short root router plus only justified scoped files; preserves or links existing CLAUDE.md; includes Rust and migration validation gates.",
      "files": []
    },
    {
      "id": 3,
      "prompt": "We maintain a skill library with skills/, registry/, scripts/, and generated docs. Generate the AGENTS.md files that make sense for this repo and explain why this is not just a specs workflow.",
      "expected_output": "Explains that agentsmd-scaffold owns AGENTS generation while repo-agent-context-audit is for broader audit/recommendation; proposes root and scoped AGENTS files for skills, registry, and scripts with concrete validation.",
      "files": []
    }
  ]
}
brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).Stars: 322kUpdated Claude CodeCodexVerifiedInstall
brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).Stars: 198kUpdated Claude CodeCodexVerifiedInstall