autopilot

Full autonomous execution from idea to working code

Quality

51%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/autopilot/SKILL.md

Quality

Content

70%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a strong operational document: clear phase workflow with validation checkpoints, escalation handling, and concrete commands and paths. Its weaknesses are token bloat in the Workflow_Profiles and Advanced sections, and progressive disclosure — advanced detail is inlined while its only external references point to files that are not in the bundle.

Suggestions

Move the Workflow_Profiles internals (descriptor/hash/Stop-hook evidence rules, V1 deferrals) and the Cursor/team-runtime details into a reference file, keeping only selection rules and valid stage sequences inline.

Ship the referenced docs/REFERENCE.md and docs/company-context-interface.md (or drop the pointers) so navigation is resolvable.

Relocate version-specific runtime notes (Claude Code 2.1.217-2.1.219 spawn-depth defaults) to a 'deprecated/version notes' section or reference file to keep the main flow lean.

Dimension	Reasoning	Score
Conciseness	Mostly operational content, but noticeably padded in places: the Workflow_Profiles section spends dense paragraphs on descriptors, SHA-256 hashes, and Stop-hook evidence semantics, and version-pinned runtime notes ('Claude Code 2.1.217-2.1.218 defaulted...') appear outside any deprecated section. Fits anchor 3 — could be tightened, though not the severe padding of anchor 2.	3 / 5
Actionability	Concrete file paths (.omc/autopilot/spec.md, .omc/plans/ralplan-*.md), executable JSONC config blocks, real commands (omc team 1:cursor "<task>", /oh-my-claudecode:cancel), and explicit subagent types give mostly executable guidance. Minor gaps (e.g. no concrete invocation shown for Phase 0/1 Analyst/Architect expansion) keep it at anchor 4 rather than 5.	4 / 5
Workflow Clarity	Six phases are strictly sequenced ('Each phase must complete before the next begins') with explicit feedback loops (QA repeats up to 5 cycles with a 3-strike stop; validation re-runs on rejection), explicit escalation/stop conditions, and a Final Checklist including fresh test/build verification. This matches the anchor-5 shape exactly, including validation of the destructive state-cleanup step.	5 / 5
Progressive Disclosure	Sections are well delimited by tags, but heavy advanced material (Workflow Profiles internals, Cursor/team config, the 3-stage pipeline) is inlined where reference files belong, and the two pointers ('See docs/REFERENCE.md', 'docs/company-context-interface.md') are dangling — no bundle files exist. Anchor 3 fits: structure present, but references are not backed by a real bundle and bulk content should be split out; not 4 given the unresolved references.	3 / 5
	Total	15 / 20 Passed

Description

33%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description communicates the skill's essence in one line but omits everything that makes a description actionable: concrete capability verbs, a 'Use when' trigger clause, and distinguishing keywords. It reads as a tagline rather than a routing description.

Suggestions

Add an explicit 'Use when...' clause with natural trigger phrases (e.g. 'Use when the user says "build me", "create me", "autopilot", or wants end-to-end hands-off execution'), which would lift both completeness and trigger_term_quality.

Enumerate 2-4 concrete actions (e.g. 'expands a 2-3 line idea into a spec, plans, implements in parallel, QA-cycles, and multi-reviewer validates') to replace the single generic capability statement.

Add a distinguishing boundary (e.g. 'for multi-phase projects, not single fixes — use direct delegation for those') to reduce conflict risk with plan/executor skills.

Dimension	Reasoning	Score
Specificity	The description 'Full autonomous execution from idea to working code' names the domain but lists no concrete actions — no expansion, planning, parallel implementation, QA, or validation verbs. It sits at anchor 2 (domain named, actions minimal/generic), below anchor 3 which requires 1-2 named concrete actions.	2 / 5
Completeness	The 'what' is stated clearly, but there is no 'Use when...' clause or any equivalent trigger guidance, which caps completeness at 3 per the judging guidelines. It is not a 4 because the 'when' is entirely absent rather than merely imprecise.	3 / 5
Trigger Term Quality	Only 'autonomous' and the generic phrase 'from idea to working code' appear; there are no natural user phrasings (e.g. 'build me', 'end-to-end', 'from scratch') that would trigger this skill. This is above anchor 1 (not pure jargon) but below anchor 3's 'some relevant keywords'.	2 / 5
Distinctiveness Conflict Risk	'Full autonomous execution... working code' is very broad and would equally match plan/build/orchestration skills — the body itself documents overlap with 'plan', 'ralph', and direct executor delegation. High overlap risk matches anchor 2.	2 / 5
	Total	9 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: Yeachan-Heo/oh-my-claudecode
Path: skills/autopilot/SKILL.md
Commit: dc7ba1d

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.