Google Agent Development Kit is a code-first toolkit for building, evaluating, deploying, and coordinating AI agents. Developers use it to create agents and multi-agent systems with tools, orchestration, tracing, and deployment options, while the catalogue entries are example agents and skills built around the toolkit.
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.
npx skills add google/adk-docs --skill api-reference-auditgit clone --depth 1 https://github.com/google/adk-docsWrote 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.
[](https://agentmods.dev/skills/google/adk-docs/api-reference-audit)<a href="https://agentmods.dev/skills/google/adk-docs/api-reference-audit"><img src="https://agentmods.dev/badge/skills/google/adk-docs/api-reference-audit/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.
<a href="https://agentmods.dev/skills/google/adk-docs/api-reference-audit"><img src="https://agentmods.dev/badge/skills/google/adk-docs/api-reference-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00098 | $0.04046 |
| Opus 5.5 | $0.00039 | $0.01618 |
| Sonnet 5.5 | $0.00020 | $0.00809 |
| Haiku 4.5 | $0.00010 | $0.00405 |
Grade A, and why
api-reference-audit 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 17d 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.
How it starts
The opening of the file, as written. The whole thing — 295 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ADK API reference doc auditor (api-ref-audit)
This skill audits the API reference documentation in the adk-docs repository across every language SDK and reports what is out of date. It is a reporting and guidance skill: it detects drift, recommends the correct per-surface release process, and emits an executable plan only when the user explicitly asks. It does not edit files, run generators, or open PRs by default.
The audit covers three kinds of drift:
- Generated API reference docs under
docs/api-reference/that lag behind the latest SDK release. - Hardcoded version strings in prose and examples (dependency coordinates,
pom.xmlversions) that must be bumped with each Java/Kotlin release. - Code-sample drift (light flagging only): samples that reference pinned versions or obviously deprecated APIs. Deep sample verification is out of scope of this skill.
[!IMPORTANT] The maps below are hints, not ground truth. File layouts, script paths, and team processes change. At each step, verify the path or marker exists and re-derive it if a lookup fails, rather than trusting the table blindly.
Version map (in-repo version vs. upstream truth)
Each surface is one API reference output this repo ships: Python API, Python CLI, Python REST API, Python Agent Config, TypeScript, Go, Java, and Kotlin.
For each surface, read the in-repo version from the file listed, then compare it against the latest upstream release and package registry.
| Surface | In-repo version marker | Upstream repo | Package registry |
|---|---|---|---|
| Python API | docs/api-reference/python/_static/documentation_options.js (VERSION:) |
google/adk-python |
PyPI google-adk |
| Python CLI | docs/api-reference/cli/_static/documentation_options.js (VERSION:) and docs/api-reference/cli/index.html ("ADK X.Y.Z") |
google/adk-python |
PyPI google-adk |
| Python REST API | docs/api-reference/rest/openapi.json (info.version) |
google/adk-python |
PyPI google-adk |
| Python Agent Config | no version marker; diff the type set in agentconfig/index.html against upstream AgentConfig.json |
google/adk-python (same release as Python API/CLI/REST) |
n/a |
| TypeScript | docs/api-reference/typescript/variables/version.html (the version const); cross-check docs/api-reference/typescript/index.html (<title> "... - vX.Y.Z") |
google/adk-js (monorepo; see note) |
npm @google/adk (authoritative) |
| Go | external, hosted on pkg.go.dev; verify links in docs/api-reference/index.md |
google/adk-go |
pkg.go.dev |
| Java | docs/api-reference/java/index.html (<title> "Maven Parent POM X.Y.Z API") |
google/adk-java (authoritative) |
Maven Central com.google.adk:google-adk (lags; cross-check only) |
| Kotlin | docs/api-reference/kotlin/index.html (version string near the header) |
google/adk-kotlin (authoritative) |
Maven Central com.google.adk:google-adk-kotlin-core (lags; cross-check only) |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 17d ago Changed · +2 lines b9f7bd688fdf
- 1mo ago First seen · 293 lines · 98 tokens per session scan A 5bbe52f3aeef
api-reference-audit is a skill published in the GitHub repository google/adk-docs (1,524 stars, last pushed yesterday), licensed Apache-2.0. It adds 98 tokens to every session and 4,046 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
asplos-writing-style
Use when drafting or revising ASPLOS prose — engineering the first two pages that rapid review actually reads, stating the cross-layer insight as one quotable sentence, structuring per-layer mechanism sections, writing bounded claims, and fitting everything into 11 self-contained pages that include figures and…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…