CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

cy-loop-tasks

Run a requested Compozy spec delivery loop through task checkpoints, scoped QA, review, and PR CI.

Quality

83%

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

92%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.

A tightly written orchestration body that routes by detected phase, owns validation through explicit completion and failure rules, and pushes all procedural detail into real, one-level-deep bundle files. The only meaningful gap is a few policy-heavy passages (evidence ownership, memory protocol) that would benefit from one concrete inline example.

Suggestions

Add one inline example of a checkpoint commit message and a filled iteration-summary line so the commit/push and evidence rules are copy-paste concrete.

Show a sample detect-phase.py output line next to the routing table so the mapping from detector output to procedure is unambiguous on first use.

Give a one-line concrete instance of 'meaningful evidence' in the evidence-ownership table (e.g. what a task-owned live/visual acceptance record looks like).

DimensionReasoningScore

Conciseness

The ~75-line body is dense with zero padding: no explanation of concepts Claude already knows, every section carries operational rules (e.g. "Its `git add -A` contract is usable only when every dirty path belongs to this checkpoint"), and detail is pushed to bundle files. It fully matches the 'lean and efficient; every token earns its place' anchor.

5 / 5

Actionability

Concrete, runnable commands are present (`python3 .agents/skills/cy-loop-tasks/scripts/detect-phase.py <slug>`, named state-writer scripts) and the detector-output-to-procedure routing table is exact. It falls short of 5 because parts of the evidence-ownership and memory sections are policy statements with no inline example (e.g. no sample checkpoint message or iteration-summary rendering).

4 / 5

Workflow Clarity

The loop is explicitly sequenced — resolve inputs once, detect phase, load only the matching procedure, complete owned work, record evidence, advance state, emit the summary, then re-detect — with explicit validation gates ("Do not write final iteration state or print a completion summary for an intermediate failure", Phase E completion criteria, and a dedicated recovery-loop reference for error diagnosis).

5 / 5

Progressive Disclosure

The body is a pure routing overview with clearly signaled, conditionally loaded, one-level-deep references; all referenced files (phase-*.md, memory-protocol.md, stacked-prs.md, etc.) and assets exist, and spot-checking confirms the referenced files are substantive rather than pointer stubs.

5 / 5

Total

19

/

20

Passed

Description

66%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.

A specific, domain-anchored description that clearly states what the skill does, but it omits any 'when to use' trigger guidance, which limits both completeness and discoverability. Adding an explicit activation clause with natural trigger phrases would lift it into the top tier.

Suggestions

Append an explicit trigger clause, e.g. 'Use when the user asks to run, deliver, or continue a Comozy spec through its task checkpoints, QA, review, and PR CI.'

Include natural synonyms users might say (e.g. 'delivery workflow', 'spec tasks', 'ship the spec') so the description triggers on looser phrasings than 'spec delivery loop'.

Distinguish the tasks-mode trigger from the free-mode sibling skill in the description to reduce overlap risk within the cy-loop family.

DimensionReasoningScore

Specificity

The description names the domain ("Comozy spec delivery loop") and several concrete actions ("task checkpoints, scoped QA, review, and PR CI") rather than generic language. It stays at 4 rather than 5 because the actions are one compressed chain and some capabilities (free-mode slices, stacked PRs, delegation) are never surfaced.

4 / 5

Completeness

The 'what' is clear ("Run a requested Comozy spec delivery loop through task checkpoints, scoped QA, review, and PR CI"), but there is no 'Use when...' clause or equivalent explicit trigger guidance, which caps completeness at 3 per the rubric guidelines.

3 / 5

Trigger Term Quality

Terms like "spec delivery", "task checkpoints", "QA", "review", and "PR CI" are natural phrases a user in this repo would say when requesting this skill. It misses 5 because there are no synonyms or looser phrasings (e.g. "delivery workflow", "spec tasks") to widen the trigger net.

4 / 5

Distinctiveness Conflict Risk

The "Comozy" brand name and spec-delivery framing carve out a clear niche with minimal generic conflict. It is not 5 because the body reveals a sibling 'free' mode / cy-loop family of skills whose scope could overlap with this description.

4 / 5

Total

15

/

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

compozy/compozy

Path

.agents/skills/cy-loop-tasks/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.