code-review-graph

Token-efficient code review using Tree-sitter AST graphs and MCP. Cuts AI token usage on large codebases by computing the blast radius of changes instead of reading entire codebases. Uses a SQLite graph database for structural analysis.

Quality

72%

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 ./.agents/skills/code-review-graph/SKILL.md

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is highly actionable with concrete, executable commands and clear sequenced workflows that include verification checkpoints. Its weaknesses are mild verbosity in the marketing-style framing and a monolithic structure with no progressive disclosure via separate reference files.

Suggestions

Tighten the overview and remove the illustrative token-savings table / "Quality angle" aside, or move them into a separate reference, to improve token efficiency.

Split long reference-style sections (Alternatives Comparison, Integration with AG Kit, full workflow details) into one-level-deep reference files and link to them from SKILL.md to improve progressive disclosure.

Trim redundant framing sentences (e.g., repeated token-savings claims in the intro and Best Practices) so every remaining token earns its place.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with concrete commands and tables, but carries padded marketing prose ("Cut AI token usage substantially on large codebases") and an illustrative token-savings table plus a "Quality angle" aside that add length without new actionable information. It is not a 3 because some passages could be tightened, and not a 1 because it largely avoids explaining concepts Claude already knows.	2 / 3
Actionability	Throughout it gives fully executable, copy-paste-ready commands ("pipx install code-review-graph", "code-review-graph build", "code-review-graph watch", "code-review-graph install --platform cursor") with concrete config examples. It is not a 2 because the guidance is complete and specific rather than pseudocode or missing key details.	3 / 3
Workflow Clarity	Installation is a clearly numbered Step 1–5 sequence with an explicit verification checkpoint (Step 5: run `/mcp` and confirm), and the Bootstrap Protocol adds Step 1–4 with confirmation gates ("Never install or run build without confirmation"). It is not a 2 because validation checkpoints and feedback gates are explicit rather than implicit or missing.	3 / 3
Progressive Disclosure	The skill is well sectioned but monolithic: no bundle files exist in references/, scripts/, or assets/, and content that could live in separate files (Alternatives Comparison, Integration with AG Kit, detailed workflow examples) is all inline. It is not a 3 because there are no clearly signaled one-level-deep references, and not a 1 because section organization is reasonable rather than a disorganized wall of text.	2 / 3
	Total	10 / 12 Passed

Description

67%

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 specific and distinctive, clearly conveying what the skill does through concrete mechanisms. Its main weakness is the lack of an explicit "Use when..." trigger clause in the description field itself, which caps completeness, and a slightly technical tone that narrows natural trigger coverage.

Suggestions

Add an explicit "Use when..." trigger clause to the description field (e.g., "Use when reviewing code in large codebases or monorepos") so the what-and-when requirement is satisfied within the description.

Soften jargon ("blast radius", "AST graphs") or pair it with plain-language equivalents ("impact of changes", "code structure") to broaden natural trigger terms users would actually say.

Consider covering a couple more everyday variations like "monorepos", "dead code", or "refactoring preview" in the description to improve trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Names several concrete mechanisms and actions — "computing the blast radius of changes", "Cuts AI token usage", "Uses a SQLite graph database", "Tree-sitter AST graphs and MCP" — matching the anchor that lists multiple specific concrete actions. It is not a 2 because the actions are concrete and varied rather than naming only a domain plus a few actions.	3 / 3
Completeness	It clearly answers "what" the skill does but "when" is only implied ("on large codebases") with no explicit "Use when..." trigger clause in the description field. Per the judging guideline, a missing explicit trigger guidance caps completeness at 2; it is not a 1 because the "what" is strong.	2 / 3
Trigger Term Quality	It includes some natural terms users would say ("code review", "large codebases", "token usage") but leans technical with jargon like "blast radius" and "AST graphs", and omits common variations. It is not a 3 because coverage of everyday user phrasing is incomplete, and not a 1 because the relevant keywords it does include are genuinely natural.	2 / 3
Distinctiveness Conflict Risk	The combination of "Tree-sitter AST graphs", "SQLite graph database", "blast radius", and "MCP" defines a clear niche unlikely to trigger for other skills. It is not a 2 because the distinctive mechanisms make overlap with similar skills unlikely.	3 / 3
	Total	10 / 12 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
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: vudovn/ag-kit
Path: .agents/skills/code-review-graph/SKILL.md
Commit: 04514f0

Reviewed: about 12 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.