component.skills.bundle-and-activation-beacons.hooks — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited component.skills.bundle-and-activation-beacons.hooks (Hook) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
aoa-skills is the AoA skill system: the canon, tooling, review surfaces, and portable export path for self-contained skill bundles a local coding agent can inspect, execute, validate, export, and improve.
A skill is a bounded workflow object. It is not a loose prompt, not project-local shorthand, and not a hidden shortcut. The system exists so AoA can grow reusable execution ability without hiding ownership, evidence, activation boundaries, or downstream adoption truth.
Use this README as the public front door. When work becomes agent-operational, skill-local, mechanical, generated, exported, or status-bearing, follow the linked owner surface instead of expanding this page.
Current release: v0.4.0. See CHANGELOG for release notes.| Function | Surface |
|---|---|
| Repository authority boundary | CHARTER |
| Skill-layer system form | DESIGN |
| Agent-facing guidance form | DESIGN.AGENTS |
| Agent route law and local checks | AGENTS |
| Technical source/generated/export model | ARCHITECTURE |
| Public skill catalog | SKILL_INDEX |
| Canonical skill bundles | skills |
| Owner-local skill mechanics | mechanics |
| Derived reader surfaces | generated |
| Portable skill export and Spark agent lane | .agents |
This repository is strongest when it turns a repeated way of working into a bounded, reviewable execution object. It is weakest when it tries to become sibling repository doctrine, proof doctrine, route law, memory, runtime, scenario composition, or downstream adoption truth.
Read only what matches your entry need.
| Need | Route |
|---|---|
| Shortest honest overview | this README, then CHARTER, DESIGN, SKILL_INDEX, and ARCHITECTURE |
| Agent editing route | AGENTS, then the nearest nested AGENTS.md |
| Skill source topology | skills |
| Boundary with sibling owner repos | LAYER_POSITION |
| Runtime use path | RUNTIME_PATH |
| Public status and evaluation | PUBLIC_SURFACE and EVALUATION_PATH |
| Install, export, and downstream refresh | INSTALL_AND_PROFILES and COMPONENT_REFRESH_LAW |
| Current direction | mechanics/ROADMAP.md, then the nearest mechanic ROADMAP.md; QUESTBOOK.md |
| AoA center requests into this repo | mechanics/OWNER_REQUEST_RECEIPTS.md |
For package-deep questions, start from mechanics and continue to the nearest mechanic package. The root README should not become the package atlas.
Canonical bundles live under skills:
| Lane | Use for |
|---|---|
| core/engineering | reusable engineering workflows such as change protocol, ADRs, source truth, TDD, contracts, invariants, ports, and boundaries |
| core/session-growth | reviewed session-growth workflows such as harvest, route forks, repair, progression, quest harvest, closeout bridge, automation scan, and summon |
| risk | explicit risk guards for approval, dry run, infra changes, stack bring-up, and sanitized sharing |
| project | thin project overlays that adapt reusable workflows to named owner families |
A bundle owns its SKILL.md, techniques.yaml, and bundle-local support artifacts. Mechanics explain movement around the canon; they do not contain canonical skills.
mechanics holds owner-local movement around the skill canon: method growth, recurrence, checkpoint carry, quest integration, audit, release support, antifragility, RPG reader surfaces, experience, and boundary bridges.
Two currently important route anchors:
keeps requested Agon workflow candidates bounded before any skill landing.
is the generated companion for that bridge.
Generated and exported surfaces are companions. Source bundles, source config, builders, review records, mechanics packages, and owner repositories keep meaning.
Use AGENTS and the nearest local AGENTS.md for executable validation routes. For release and public-claim checks, start from RELEASING and the relevant release-support or audit surface.
Grow the skill canon by making the next execution route clearer.
Add bundles, mechanics, generated companions, examples, tests, and exports only where they improve reviewability and preserve owner boundaries. When detail belongs to a technique, proof surface, playbook, route, memory object, runtime body, downstream repository, roadmap, landing log, changelog, quest, decision record, or agent route card, route it there.
Apache-2.0
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.