cog-review

Cog architecture and cross-cutting review guidelines

Quality

60%

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 ./.opencode/skills/cog-review/SKILL.md

Quality

Content

72%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 concise, well-structured, and packed with concrete repo-specific guidance that respects Claude's competence. Its main weakness is that it reads as a reference checklist rather than a sequenced review workflow with validation checkpoints.

Suggestions

Convert 'What to watch for' into a short sequenced review checklist with explicit validation steps (e.g. 'For each change: verify VERSION.txt matches Cargo.toml, confirm schema and Go code are in sync, check both Rust and Python IPC sides are updated').

Add a one-line 'When to use this skill' trigger at the top of the body to mirror the missing frontmatter guidance.

Consider a quick 'verify before approving' checklist item so reviewers have an explicit feedback loop for risky changes.

Dimension	Reasoning	Score
Conciseness	The body is efficiently organized into tight sections of repo-specific facts Claude would not already know, with only minor soft explanatory clauses ('Changes to build artifacts need to account for this') that could be trimmed, matching the score-4 anchor.	4 / 5
Actionability	As an instruction-only review skill it provides concrete, specific guidance via exact paths and commands (e.g. 'mise run docs:llm', 'pkg/config/data/config_schema_v1.0.json'), matching the score-4 anchor of mostly executable guidance with minor gaps.	4 / 5
Workflow Clarity	The body is a well-structured reference (overview, patterns, concerns, watch-for) rather than a sequenced multi-step workflow, and the 'What to watch for' checklist lacks explicit validation checkpoints, fitting the score-3 anchor of steps listed but checkpoints missing or implicit.	3 / 5
Progressive Disclosure	At ~46 lines with no need for external references, the body is organized into clear, well-signaled sections (Component overview, Key design patterns, Cross-cutting concerns, What to watch for) with easy navigation, meeting the under-50-lines exception that allows a score of 5.	5 / 5
	Total	16 / 20 Passed

Description

48%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 domain-specific and reasonably distinct, but it lacks concrete capability actions and any explicit 'Use when...' trigger guidance, capping completeness and specificity. Adding trigger phrases and naming concrete review actions would materially improve it.

Suggestions

Add an explicit 'Use when...' clause naming the situations that should trigger this skill (e.g. 'Use when reviewing changes that cross language boundaries or touch core Cog architecture').

Replace the generic 'review guidelines' with 1-2 concrete actions the skill performs (e.g. 'flags breaking changes to the predictor interface and cog.yaml schema').

Include natural user-facing trigger terms alongside the jargon (e.g. 'PR review', 'architecture change') so users would naturally invoke the skill.

Dimension	Reasoning	Score
Specificity	The description names the domain ('Cog architecture and cross-cutting review') but the only action is the generic 'review guidelines', matching the score-2 anchor where the domain is named but actions are minimal or generic.	2 / 5
Completeness	It gives a clear 'what' (architecture and cross-cutting review guidelines) but no 'Use when...' trigger clause, so per the guideline capping completeness at 3 for missing explicit trigger guidance it lands on the score-3 anchor.	3 / 5
Trigger Term Quality	Terms like 'Cog', 'architecture', 'review', and 'cross-cutting' are relevant to the domain but jargon-heavy and miss the natural phrasing a user would say, fitting the score-3 anchor of relevant keywords missing common variations.	3 / 5
Distinctiveness Conflict Risk	'Cog architecture' is a specific niche with minimal conflict risk outside that domain, though it could overlap with other code-review skills, matching the score-4 anchor of mostly distinct with minor overlap risk.	4 / 5
	Total	12 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: replicate/cog
Path: .opencode/skills/cog-review/SKILL.md
Commit: ec8082f

Reviewed: 5 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.