CtrlK
BlogDocsLog inGet started
Tessl Logo

autopilot

[OMX] Canonical autonomous orchestrator: $deep-interview -> $ralplan -> $ultragoal

56

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/oh-my-codex/skills/autopilot/SKILL.md

The canonical home for this skill is autopilot in Yeachan-Heo/oh-my-codex

SKILL.md
Quality
Evals
Security

Quality

Content

73%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 well-structured orchestration policy with an exceptionally clear workflow: mandatory phase order, explicit transition validation gates, a review loopback, recovery contracts, and a final checklist. Actionability is strong thanks to the concrete state schema and CLI command, though repeated restatements of the same recovery/authority constraints and the inlined state detail cost it some conciseness and structure points.

Suggestions

Consolidate the recovery/fail-open and host-receipt constraints, which currently appear in slightly different wording in Execution_Policy, State_Management, and Recovery_Contract, into one authoritative section referenced by the others.

State the concrete storage location of handoff artifacts and the exact CLI command for the review/QA loopback state update, mirroring the copy-paste pattern used for the initial state write.

Consider moving the full State_Management JSON schema and CLI payload into a reference file, keeping SKILL.md at a lean overview with the key transition gates inline.

DimensionReasoningScore

Conciseness

The body avoids teaching known concepts and is dense with operative policy, but the recovery/fail-open and "no host-issued receipt" constraints are restated across Execution_Policy, State_Management, and Recovery_Contract, and the chain itself is repeated five times, so it could be tightened.

3 / 5

Actionability

Concrete, copy-paste-ready guidance is present — the full state JSON schema, the `omx state write` command with its complete payload, named transition-gate artifacts, `$cancel`, and per-phase resume semantics — with only minor gaps such as unspecified artifact storage paths and no exact command for loopback state updates.

4 / 5

Workflow Clarity

The multi-stage process has a mandatory numbered phase order, explicit validation gates at every transition (durable handoff artifacts, approvals, completion receipts, fresh verification evidence), a review/QA feedback loop back to ralplan, cancellation and recovery paths, and a Final Checklist — matching the top anchor.

5 / 5

Progressive Disclosure

There are no bundle files, and the single file is well organized into clearly tagged sections that make navigation easy; however, at ~120 lines with inlined state schemas and CLI payloads, some State_Management detail could arguably live in a reference file, leaving minor organization gaps.

4 / 5

Total

16

/

20

Passed

Description

53%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 states a clear and distinct "what" — a canonical orchestrator for the mandatory deep-interview → ralplan → ultragoal chain — but omits any "when to use" guidance and leans on internal jargon ([OMX], $-commands) rather than natural user trigger phrases. It would benefit from a sentence explaining what the chain accomplishes and when to reach for it.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user asks for end-to-end hands-off delivery from an idea, issue, or requirements seed, or explicitly invokes $autopilot."

Briefly state what each stage contributes (e.g. "clarifies requirements, builds a consensus plan, then executes and verifies it") so the "what" is understandable without knowing the sub-skills.

Drop or de-emphasize ecosystem jargon like "[OMX]" and "Canonical" in favor of natural keywords such as "autonomous delivery", "end-to-end", and "hands-off".

DimensionReasoningScore

Specificity

The description names the domain ("Canonical autonomous orchestrator") and three concrete stage commands ("$deep-interview -> $ralplan -> $ultragoal"), but it never states what those stages actually do, so action coverage is incomplete rather than having only minor gaps.

3 / 5

Completeness

It answers "what" (a canonical orchestrator that runs the named three-stage chain) but contains no "Use when..." clause or equivalent explicit trigger guidance, which caps completeness at 3. Not 2, because the "what" is clearly stated rather than vague or absent.

3 / 5

Trigger Term Quality

Some relevant keywords are present ("autonomous", "orchestrator", the stage names), but the natural phrases a user would say — e.g. "end-to-end delivery", "hands-off", "from requirements to done" — are missing, and "[OMX]", "Canonical", and the $-prefixed commands are internal jargon rather than user-sayable triggers.

3 / 5

Distinctiveness Conflict Risk

The fixed three-stage chain carves a clear niche with minimal conflict risk against unrelated skills, but the description's trigger terms are exactly the names of the sub-skills it orchestrates ($deep-interview, $ralplan, $ultragoal), leaving minor overlap risk with those closely related skills.

4 / 5

Total

13

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
Yeachan-Heo/oh-my-codex
Reviewed

Table of Contents

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.