skill-anatomy
Use when authoring or updating a cadence workflow — contributing one upstream, or adding your own alongside it.
Skill by arbor-ai-inc
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/arbor-ai-inc-skill-anatomy
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: productivity, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/skill-anatomy
What's inside
- Skill name
skill-anatomy- Description
- Use when authoring or updating a cadence workflow — contributing one upstream, or adding your own alongside it. Covers the canonical/adapter split, subagent format, and the presubmit guardrails.
- License
- Apache-2.0
- Instructions
- 18 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
<!-- GENERATED by scripts/gen_adapters.py from skills/skill-anatomy/SKILL.md.
Do not edit: run the generator instead. The canonical workflow is
reference/skill-anatomy.md; this file only routes an agent to it. -->
Follow the canonical procedure in `.cadence/reference/skill-anatomy.md` exactly.
**One canonical statement per workflow in `reference/`; every agent-facing file routes to it.** That split is load-bearing, not tidiness: a 900-line doc cannot sit in every session's context, and a rule in two places has already begun to drift.
An adapter refers to its canonical doc through `${CLAUDE_PLUGIN_ROOT}` — a repo-relative path resolves to nothing from an installed plugin.
**Never pin `model:` in anything shipped.** It silently overrides the user's choice for a role they are paying for. `tests/check_agent_skills.py` fails on a pin.
A subagent prompt carries role mechanics — who it is, what it may touch, the output and verdict format. The rules it grades by belong in the canonical doc.
Name the command, or name where the command is configured, never neither. Hardcoding one project's test invocation makes a workflow unusable elsewhere; "run the tests" makes it unusable anywhere.
Verify with `python3 tests/check_agent_skills.py`, and break any new check once to confirm it fires.From arbor-ai-inc/cadence, Apache-2.0.
More from arbor-ai-inc/cadence
- 0.565score
git-pr-workflow
arbor-ai-inc
Use when creating a branch, committing, opening a pull request, driving an automated review through to settled, choosing a merge strategy, or cleaning up…
SkillAFreshness: AClaude Code +9 - 0.565score
spec-pipeline
arbor-ai-inc
Orchestrates the full spec review pipeline to READY, consulting the author only for decisions.
SkillAFreshness: AClaude Code +9 - 0.565score
code-review
arbor-ai-inc
Use when a feature branch needs adversarial review before PR: the configured reviewer, resolve findings, re-run lint and tests, loop to LGTM.
SkillAFreshness: AClaude Code +9 - 0.565score
review-spec
arbor-ai-inc
Runs one spec-pipeline review round — Gate 1 (product) or Gate 2 (eng design) — by invoking the matching reviewer subagent.
SkillAFreshness: AClaude Code +9 - 0.565score
decision-fanout
arbor-ai-inc
Use when an autonomous implementation run hits a decision with more than one defensible answer and the project set [fanout].enabled = true: build each option…
SkillAFreshness: AClaude Code +9 - 0.565score
draft-plan
arbor-ai-inc
Drafts design.md for a spec, against the project's architectural principles.
SkillAFreshness: AClaude Code +9 - 0.565score
execute-issue
arbor-ai-inc
Use when picking up a tracker issue (e.g.
SkillAFreshness: AClaude Code +9 - 0.559score
explain-plain
arbor-ai-inc
Use when explaining or summarizing any source — a tracker issue, a file, a diff, a chat thread — in plain, everyday English for a non-technical reader, in one…
SkillAFreshness: AClaude Code +9 - 0.529score
test-authoring
arbor-ai-inc
Use when planning the tests for a spec, when adding or changing an integration, component behavioral, smoke or E2E test or shared fixture, or when a change…
SkillAFreshness: AClaude Code +9 - 0.524score
init
arbor-ai-inc
Use when setting up cadence in a project for the first time, or when asked where cadence's config and templates are.
SkillBFreshness: BClaude Code +9 - 0.524score
retro-synthesis
arbor-ai-inc
Use when the retros pending/ directory holds at least the configured number of fragments and the batch should become guidance changes, or when a guidance doc…
SkillBFreshness: BClaude Code +9 - 0.519score
code-review-and-quality
arbor-ai-inc
Use when reviewing a diff, hardening code, or preparing a change for merge — the broad quality pass across correctness, tests, architecture, security and…
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 8, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/arbor-ai-inc/cadence
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from arbor-ai-inc
- 0.565score
git-pr-workflow
arbor-ai-inc
Use when creating a branch, committing, opening a pull request, driving an automated review through to settled, choosing a merge strategy, or cleaning up…
SkillAFreshness: AClaude Code +9 - 0.565score
code-review
arbor-ai-inc
Use when a feature branch needs adversarial review before PR: the configured reviewer, resolve findings, re-run lint and tests, loop to LGTM.
SkillAFreshness: AClaude Code +9 - 0.565score
execute-issue
arbor-ai-inc
Use when picking up a tracker issue (e.g.
SkillAFreshness: AClaude Code +9 - 0.565score
decision-fanout
arbor-ai-inc
Use when an autonomous implementation run hits a decision with more than one defensible answer and the project set [fanout].enabled = true: build each option…
SkillAFreshness: AClaude Code +9 - 0.565score
draft-plan
arbor-ai-inc
Drafts design.md for a spec, against the project's architectural principles.
SkillAFreshness: AClaude Code +9 - 0.565score
review-spec
arbor-ai-inc
Runs one spec-pipeline review round — Gate 1 (product) or Gate 2 (eng design) — by invoking the matching reviewer subagent.
SkillAFreshness: AClaude Code +9
Cite this page
skill-anatomy. Agent Mart. https://agentmart.dev/skills/arbor-ai-inc-skill-anatomy (last updated 2026-10-24)