agent-code-review-swarm

Agent skill for code-review-swarm - invoke with $agent-code-review-swarm

2.64x

Quality

37%

Does it follow best practices?

Impact

98%

2.64x

Average score across 3 eval scenarios

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/agent-code-review-swarm/SKILL.md

Quality

Content

46%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 feature-catalog style document rich with executable gh/ruv-swarm commands but lacking a single clear end-to-end review workflow with validation checkpoints. It is verbose with duplicated sections and inlines content that belongs in separate reference files, some of which are referenced but missing.

Suggestions

Consolidate the duplicated agent descriptions ("Core Features" vs "Review Agents") and merge the two "Advanced Features" sections into one to cut length and redundancy.

Add one explicit numbered end-to-end workflow (fetch PR diff -> run agents -> validate/summarize findings -> post review) with a validation checkpoint before calling `gh pr review --request-changes` or `--approve`.

Move the per-agent check lists, YAML configs, and comment templates into real reference files under references/ and link to them, or remove the broken See-also links to swarm-pr.md and workflow-automation.md.

Dimension	Reasoning	Score
Conciseness	The ~540-line body is noticeably verbose with duplicated coverage: agents are described in "Core Features" and again in "Review Agents", and there are two separate "Advanced Features" sections plus overlapping "Advanced Review Features" and "Examples". This matches the anchor for several unnecessary/padded sections, and is above score 1 because it is not pure concept explanation.	2 / 5
Actionability	It provides many concrete, copy-pasteable bash commands using gh CLI and `npx ruv-swarm github review-*`, plus config and template examples. This matches the anchor for mostly executable guidance with minor gaps; it is not score 5 because the ruv-swarm subcommands are presented without confirming they are real/installed.	4 / 5
Workflow Clarity	Steps exist (init swarm, run per-agent reviews, post comments) but they are scattered across feature sections rather than a single sequenced workflow, and there are no validation/verification checkpoints before posting reviews or approving PRs. Per the rubric, batch/outward-facing operations without validation cap workflow clarity at 3.	3 / 5
Progressive Disclosure	No bundle files exist (references/scripts/assets absent), yet all agent check lists, config YAML, comment templates, and CI workflows are inlined into one monolith. The only file references (swarm-pr.md, workflow-automation.md) appear as a buried "See also" line and point to non-existent files, matching the anchor for inlined content and buried/broken references.	2 / 5
	Total	11 / 20 Passed

Description

28%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 frontmatter description is a near-placeholder that names the skill and its invocation syntax but does not explain what it concretely does or when to use it. It fails to provide natural trigger phrases or an explicit Use-when clause.

Suggestions

Rewrite the description to state concrete capabilities, e.g. "Runs specialized AI agents (security, performance, architecture, style) to review GitHub pull requests and post findings via gh CLI."

Add an explicit "Use when..." clause naming natural trigger phrases such as "code review", "review a PR", "security review of a pull request", or "multi-agent code analysis".

Drop the "invoke with $agent-code-review-swarm" meta syntax from the description; keep it concise and in third person.

Dimension	Reasoning	Score
Specificity	The description only names the domain ("Agent skill for code-review-swarm") with no concrete review actions; "invoke with $agent-code-review-swarm" is invocation syntax, not a capability. It sits at the anchor for naming a domain with minimal/generic actions, and is not score 1 only because it does identify the domain.	2 / 5
Completeness	It has only a vague "what" ("Agent skill for code-review-swarm") and no "when"/"Use when..." clause at all, matching the anchor for a vague what with no when. The missing explicit trigger guidance also caps completeness per the guidelines.	2 / 5
Trigger Term Quality	The only keyword is the product slug "code-review-swarm"; natural user phrases like "code review", "review a PR", or "security analysis" are absent, and the wording is meta/invocation language. This matches the anchor for one or two generic keywords missing the natural phrases users say.	2 / 5
Distinctiveness Conflict Risk	The slug "code-review-swarm" points at a specific niche, so it is somewhat distinguishable, but the description is too sparse to clearly differentiate it from other code-review skills, leaving moderate overlap risk.	3 / 5
	Total	9 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (543 lines); consider splitting into references/ and linking	Warning
relative_links	Relative link issues: 4 missing	Warning

	Total	14 / 16 Passed

Repository: ruvnet/ruflo
Path: .agents/skills/agent-code-review-swarm/SKILL.md
Commit: fa13ee4

Reviewed: about 4 hours 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.