Learn how to author a third-party gbrain skillpack through a reference example. It explains required files, ten validation criteria, tests, and evaluation coverage, helping authors understand what to check before packaging and publishing.
Reference skill demonstrating the 10/10 skillpack contract. Adds a "what does this skillpack do" answer to the user's agent.
Review the instructions the agent will receive before you install.
This is the canonical reference for a third-party gbrain skillpack. Read its tree once and you know how to author one.
When the user asks how third-party gbrain skillpacks work, this skill points them at the four artifacts every pack ships:
skillpack.json — declares pack metadata + which artifacts the
doctor scores (skills, unit_tests, e2e_tests, llm_evals, routing_evals,
runbooks, changelog).skills/<name>/SKILL.md — frontmatter (name, description, mutating,
triggers) plus markdown body. Agents route on triggers:; the body
is the in-context instruction set.runbooks/bootstrap.md — agent-readable post-scaffold steps. gbrain
DISPLAYS this after scaffold lands; the agent walks per-step at
its own discretion. No auto-executor (codex T1 supply-chain hardening).CHANGELOG.md — Keep-a-Changelog shape. The doctor's changelog_ present_and_current dimension fails if there's no ## [<version>]
entry matching the manifest's version.Ten binary dimensions, split into:
Core (5; must all pass to publish at any tier):
manifest_valid — schema-validates skillpack.jsonskills_have_skill_md — every listed skill has SKILL.md with
name / description / triggersrouting_evals_present — each skill has routing-eval.jsonl with
= 5 intents
skills_have_unique_triggers — MECE at the pack levelchangelog_present_and_current — CHANGELOG entry for current versionQuality badges (5; earn for tier eligibility):
unit_tests_present — manifest.unit_tests matches >= 1 filee2e_tests_present — manifest.e2e_tests matches >= 1 filellm_eval_present — *.judge.json with >= 3 casesbootstrap_runbook_present — non-empty runbooks/bootstrap.mdlicense_present — LICENSE / LICENSE.md / LICENSE.txt non-emptyTier eligibility:
endorsed — all 10 (gates: Garry's endorsements.json overlay
in the registry)community — all 5 core + >= 3 of 5 badges (default tier on PR
merge)experimental — all 5 core + < 3 badgesblocked — any core failsRun gbrain skillpack doctor <pack-dir> to see exactly which
dimensions a candidate pack passes and the paste-ready fix for each
failure. --fix --yes auto-scaffolds the dimensions flagged
auto_fixable: true (routing-eval stubs, CHANGELOG entries,
license stub, bootstrap stub, test stubs).
The triggers above route any "what is a skillpack" / "how do third-
party packs work" user phrasing to THIS skill. The agent reads the
markdown body, then either answers the user's question directly or
calls gbrain skillpack info <name> / search <query> for live
registry data.
test/example.test.ts — unit test that imports the skill helper
(stub; replace with real assertions).e2e/example.e2e.test.ts — integration test gated on DATABASE_URL.evals/reference-pack.judge.json — LLM-judge eval scoring this
skill's output against the "does it actually teach the contract"
bar across happy-path / edge / failure-mode cases.skills/reference-pack/routing-eval.jsonl — five phrasings users
ask, all mapped to expected_skill: reference-pack.Before pushing, the publisher runs:
gbrain skillpack doctor . --quick --json
gbrain skillpack pack
The first hits the rubric and prints paste-ready fixes for any failed
dimension. The second emits reference-pack-<version>.tgz with a
deterministic SHA-256 the publisher submits to the registry PR.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Check research claims cited in books, articles, or conversations by tracing the original paper, methods, underlying data, and independent replication. Save a sourced assessment explaining whether the claim holds up and what remains uncertain.
Explore old files and email archives to rediscover personal writing, ideas, and meaningful conversations. Review selected items one at a time, save worthwhile material as organized notes, and track progress within explicitly permitted folders.
Restructure raw article text stored in gbrain into summaries, exact quotes, key insights, and links to related pages. Preserve the original text and connect the article's significance to projects and interests already recorded in gbrain.
Present clear choices when a task needs a human decision, then pause execution until the user replies. This reusable interaction pattern supports filing decisions, priority selection, and confirmation before destructive actions, with an option to decline.
Collect public posts across a blog, newsletter, or feed archive and save clean Markdown pages for each article. The workflow avoids duplicates, supports resumed runs, repairs empty results, and hands articles to a separate enrichment skill.
Create a detailed chapter-by-chapter reading companion from an EPUB or PDF, connecting the book’s ideas to experiences and words saved in your personal knowledge store. The result highlights parallels without prescribing actions.