{
  "markdown": "# agent-platform-skills\n\nSkills for building on **Gemini Enterprise Agent Platform** — the parts that are easy\nto get wrong and slow to diagnose: which registration mode can render A2UI, what\nauthenticates on each path, how Agent Identity changes outbound calls, how to tell a\nrejected credential from a missing permission, and which store actually keeps data\nacross restarts and users.\n\nDerived from running deployments. Where a claim was verified by observation rather\nthan taken from documentation, the skill marks it; where documentation and observed\nbehaviour disagree, both are recorded.\n\n## Skills\n\n| Skill | Use it for |\n| :--- | :--- |\n| [`agent-platform-architecture`](skills/agent-platform-architecture/SKILL.md) | **Designing.** The platform constraints that govern how agents fit together, and the six architectures they permit. One agent or several, Agent Identity or not, whether you need Agent Gateway, how to expose an agent in Gemini Enterprise. |\n| [`agent-platform-a2a`](skills/agent-platform-a2a/SKILL.md) | **Debugging.** 401 vs 403, identifying the real caller from audit logs, Agent Identity principal forms, reading permissions out of errors, and the conditions under which a broken deployment reports success. |\n| [`agent-platform-implementation`](skills/agent-platform-implementation/SKILL.md) | **Building and verifying.** The deployment runs and is quietly wrong: what actually crosses an agent-to-agent hop, why a caller invents data a sub-agent holds, guards a model cannot game, how to build an A2UI surface that survives the round trip, and the verification traps that report a score unrelated to your agent. |\n| [`agent-platform-state`](skills/agent-platform-state/SKILL.md) | **Persisting.** Where data lives and what survives: the three stores and their lifetimes, why `app:`-prefixed state does not scope on Agent Runtime, memory scopes and sharing a collection across users, and which store fits which workload. |\n\nThe split is design, build, operations, and state, so that none crowds out the\nothers: the architecture skill states the constraints and the designs they\npermit, the implementation skill covers what goes wrong once it runs without\nerroring, the debugging skill covers failure modes and how to identify them, and\nthe state skill covers where data lives and what survives a restart.\n\n## Install\n\nBoth installers read the `skills/` directory at the repo root.\n\n```bash\n# Claude Code (global)\nnpx -y skills add https://github.com/alanblythe/agent-platform-skills -y --agent claude-code -g\n\n# Project scope instead of global — drop the -g and run from the project directory\nnpx -y skills add https://github.com/alanblythe/agent-platform-skills -y --agent claude-code\n\n# Antigravity / Gemini CLI\nnpx -y skills add https://github.com/alanblythe/agent-platform-skills -y --agent antigravity -g\n```\n\nTo use it without installing, point your agent at\n[`skills/agent-platform-a2a/SKILL.md`](skills/agent-platform-a2a/SKILL.md) directly — it\nis a single self-contained Markdown file.\n\n## Layout\n\n```\nplugin.json               # Claude Code plugin manifest (duplicated in .claude-plugin/)\n.claude-plugin/plugin.json\ngemini-extension.json     # Antigravity / Gemini extension manifest\nskills/\n  README.md\n  agent-platform-architecture/\n  agent-platform-implementation/\n  agent-platform-a2a/\n    SKILL.md              # YAML frontmatter (name, description, metadata) + body\n  agent-platform-state/\n```\n\nEach skill is a directory under `skills/` containing a `SKILL.md` whose frontmatter\ncarries `name` and a `description` written as trigger phrases — that description is what\nan agent matches against when deciding whether the skill is relevant, so it names the\nsymptoms (\"A2UI renders as raw JSON\", \"401 between agents\") rather than summarising the\ncontents.\n\n## Scope\n\nDeliberately not covered: writing agent code, evaluation methodology, and scaffolding.\nThose are handled by the [`google-agents-cli`](https://github.com/google/agents-cli)\nskills, which this complements rather than replaces — install both.\n\nThe one exception is a section of `agent-platform-implementation`: where the eval\nharness reports a number that is not about your agent — a stale trace regraded as\nif it were new, a metric grading the transport your users are not served, a\nmulti-turn case whose earlier turns were narrated rather than run. Choosing metrics,\ndataset schema and the flywheel itself remain with `google-agents-cli`.\n\n## License\n\nApache-2.0\n",
  "bytes": 4471,
  "sha": "46ecdfec680227b84770050486d1b23b4315427d799a036fb769acadd997365d",
  "repo_slug": "alanblythe/agent-platform-skills",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_alanblythe_agent_platform_skills_6783724d/readme"
}