pr-review

Standardized guidelines and templates for structuring Pull Request reviews so feedback is actionable, encouraging, and extractable by the Native Edge Graph. MANDATORY ROI WARNING: Skipping the review template guarantees CI lint failure. Triggers: Reviewing a PR (yours or peer's) — structured eval metrics, graph ingestion tags, severity ladder, restates §0 merge gate, post-comment A2A commentId hand-off (reviewer→author) per guide §10, Evidence Audit + Source-of-Authority sections (template §) for substrate/runtime-AC PRs and authority-citation review-comments.

Quality

88%

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

Quality

Content

92%Weight 40%Scale 1-3

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 tight, actionable orchestration layer with a clear cycle-based workflow and an explicit validation gate, scoring well on conciseness, actionability, and workflow clarity. Its main weakness is progressive-disclosure integrity: one referenced audit file is missing and the primary guide path does not match the actual bundle root.

Suggestions

Fix or remove the dead reference: 'audits/review-cost-circuit-breaker.md' is not present in the bundle — either add the file under references/audits/ or point to one of the existing audits (decile-anchors.md, mcp-tool-description-budget.md).

Align the guide path with the bundle: cite 'references/pr-review-guide.md' (or a path that resolves from the skill root) instead of '.agents/skills/pr-review/references/pr-review-guide.md'.

Define the closure shorthand (RC2, A+FU, D+S, V-B-A) inline in one phrase or in the loaded template so the dispatch body is self-contained without the references.

Dimension	Reasoning	Score
Conciseness	The body is ~150 words and assumes Claude's competence without explaining what a PR or a memory sweep is; every line dispatches work ('use the view_file tool to read ...', 'run a ... query_raw_memories / query_summaries sweep'), matching the lean-and-efficient anchor; not a 2 because there is no padded explanation of known concepts.	3 / 3
Actionability	Gives concrete, executable dispatch: named tool ('view_file'), named files ('pr-review-guide.md', 'pr-review-template.md', 'pr-review-followup-template.md'), a call budget ('3–10-call'), and a cycle/closure decision rule; not a 2 because instructions are specific commands rather than vague description or pseudocode.	3 / 3
Workflow Clarity	Sequences a multi-step process by cycle (Cycle 1 full template, Cycle N≥2 followup delta, budgeted closure at RC2 / >24KB) with an explicit pre-scoring validation checkpoint ('before scoring, run a ... sweep ... The sweep RESULT is the V-B-A'); not a 2 because the validation gate is explicit rather than missing or merely implied.	3 / 3
Progressive Disclosure	The cycle-based dispatch to one-level-deep references is well-signaled, but scored against the actual bundle it is unreliable: 'audits/review-cost-circuit-breaker.md' does not exist in references/audits/ (only decile-anchors.md and mcp-tool-description-budget.md are present) and the guide is cited at '.agents/skills/pr-review/references/pr-review-guide.md' rather than the bundle's 'references/pr-review-guide.md'; not a 3 because easy navigation is broken by the dead and misrooted paths, not a 1 because the references are clearly signaled and not deeply nested.	2 / 3
	Total	11 / 12 Passed

Description

85%Weight 40%Scale 1-3

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 answers both 'what' and 'when' with explicit triggers and a well-defined niche, but is overloaded with internal jargon and an over-claim 'MANDATORY ROI WARNING' that crowds out natural trigger terms. It is strong on completeness and distinctiveness, weaker on trigger-term naturalness.

Suggestions

Lead with plain-language trigger phrasing (e.g., 'Use when reviewing a pull request, whether your own or a peer's') and push terms like 'Native Edge Graph', 'A2A commentId hand-off', and 'substrate/runtime-AC' into the body or references.

Drop the 'MANDATORY ROI WARNING: ... guarantees CI lint failure' over-claim from the description; it reads as a threat rather than capability information and adds buzzword noise.

Trim the section-symbol citations ('guide §10', 'template §', '#13354') from the description — they are navigation aids, not capability statements.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions such as 'structured eval metrics, graph ingestion tags, severity ladder ... post-comment A2A commentId hand-off ... Evidence Audit + Source-of-Authority sections', matching the multiple-concrete-actions anchor; it is not a 2 because the actions are individually named rather than a single vague domain claim.	3 / 3
Completeness	It clearly answers 'what' ('Standardized guidelines and templates for structuring Pull Request reviews ...') and 'when' with an explicit 'Triggers:' clause ('Reviewing a PR (yours or peer's)'), satisfying both halves; not a 2 because the trigger guidance is explicit rather than implied.	3 / 3
Trigger Term Quality	The one genuinely natural trigger is 'Reviewing a PR (yours or peer's)', but the bulk is dense internal jargon ('Native Edge Graph', 'A2A commentId hand-off', 'substrate/runtime-AC PRs', '#13354') no user would naturally say, so it has 'some relevant keywords but missing common variations' rather than full natural coverage.	2 / 3
Distinctiveness Conflict Risk	It targets a clear niche (PR review) with a distinct trigger unlikely to fire for unrelated skills, and the hyper-specific substrate/AC framing further separates it from generic skills; not a 2 because it would not plausibly overlap with sibling skills.	3 / 3
	Total	11 / 12 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: neomjs/neo
Path: .agents/skills/pr-review/SKILL.md
Commit: b4a1b58

Reviewed: 8 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.