CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

aw

Ships coding work to a tested draft PR from one opt-in entry point: detects the tier, runs Micro/Lite here, and on Full writes a short brief with sealed checks, works to them here, and converges once with a local review loop before the PR; prints a paste-ready /goal. Use when asked to "implement autonomously", "independently", "in isolation", "in a worktree", "end-to-end", "all the way to a PR", "ship this", "land this", "take care of this", "handle this without me", or /aw.

Quality

92%

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

SKILL.md
Quality
Evals
Security

Quality

Content

85%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 tightly written dispatcher: fully executable Skill() calls, exact output contracts, and a well-sequenced tier flow with validation and degradation reporting throughout. Its one real weakness is that the bundle backs almost none of its referenced files, so the progressive-disclosure structure cannot be followed from this skill directory alone.

Suggestions

Ship the referenced rule files (rules/full-tier.md, rules/lessons.md, templates/routing.rule.md) inside this skill's bundle, or inline their one-line essentials, so the routing is followable when the companion set is not installed.

Deduplicate the review-convergence policy, which currently appears in both the Routing table's per-tier rows and the Review section, keeping the full detail only in Review.

For cross-skill references like ../rules/goal-emission.md and ../rules/safety-guardrails.md, state the fallback behavior when the companion workflow skill is absent, mirroring the explicit stop-and-ask already specified for Skill("autonomous-workflow").

DimensionReasoningScore

Conciseness

The body is dense and assumes Claude's competence, delegating design rationale to references/dispatcher-rationale.md rather than inlining it, but the review-convergence policy is stated in both the Routing table and the Review section, and per-tier routing repeats across First actions, Routing, and Review. Minor trimming opportunities keep it at the efficient-with-minor-instances anchor rather than the every-token-earns-its-place anchor at 5.

4 / 5

Actionability

Guidance is fully executable: exact Skill(...) invocations (e.g. Skill("create-pr", "--no-pre-review --no-post-review")), a ready-to-run sed command for extracting the tier-table section, exact output formats for the MODE SELECTION and AW RUN COMPLETE blocks, a complete brief plus /goal example, and wrong/right disambiguation blocks. Beyond concrete commands, the worked examples make the guidance copy-paste ready with essentially no gaps.

5 / 5

Workflow Clarity

The multi-step process is clearly sequenced (First actions 0-3, tier detection, routing, review, brief/goal, follow-ups, terminal contract) with explicit validation checkpoints: the MODE SELECTION block, checks.lock sealing for resume, review-loop caps, the AW EVIDENCE block for goal judging, and a terminal contract emitted on every exit with Degraded/Needs-you reporting. Error-recovery paths (unsealed brief, unavailable companion skill, degraded tool grants named rather than skipped) are explicit, matching the top anchor.

5 / 5

Progressive Disclosure

References are clearly signaled with read-when cues ("read when the tier is Full", "read it now", "read when planning ends"), but scored against the actual bundle structure, most referenced paths — rules/full-tier.md, rules/lessons.md, templates/routing.rule.md, ../rules/goal-emission.md, ../rules/safety-guardrails.md, and the companion ../SKILL.md — are absent; only references/dispatcher-rationale.md exists, so navigation from this bundle alone breaks. The missing files are the bulk of the skill's deferred content rather than a minor gap (so not 4), but the in-body structure is well organized and existing references are one level deep (so not 2).

3 / 5

Total

17

/

20

Passed

Description

100%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 is exemplary: third person, concrete about what the skill does at each tier, and equipped with a rich set of natural trigger phrases plus the /aw command name. Both the capability set and the invocation conditions are explicit, leaving nothing to inference.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions covering the full pipeline — "detects the tier", "runs Micro/Lite here", "writes a short brief with sealed checks", "converges once with a local review loop before the PR", "prints a paste-ready /goal" — matching the comprehensive-coverage anchor rather than the several-actions-with-minor-gaps anchor at 4.

5 / 5

Completeness

It explicitly answers both questions: the "what" is concrete (ships coding work to a tested draft PR with per-tier mechanics) and the "when" is an explicit "Use when asked to..." clause with concrete trigger phrases, exactly matching the top anchor.

5 / 5

Trigger Term Quality

Natural trigger phrases are comprehensively covered with synonyms and the command name: "implement autonomously", "independently", "in a worktree", "end-to-end", "all the way to a PR", "ship this", "land this", "take care of this", "handle this without me", and "/aw".

5 / 5

Distinctiveness Conflict Risk

It carves a clear opt-in autonomous-workflow niche with distinct triggers and explicit entry-point framing; "take care of this" is mildly broad but the opt-in framing keeps conflict risk minimal, fitting the clear-niche anchor better than the minor-overlap anchor at 4.

5 / 5

Total

20

/

20

Passed

Validation

81%

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

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

relative_links

Relative link issues: 6 missing, 3 suspicious

Warning

Total

13

/

16

Passed

Repository

mthines/agent-skills

Path

skills/workflow/autonomous-workflow/aw/SKILL.md

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.