understand

Analyze a codebase to produce an interactive knowledge graph for understanding architecture, components, and relationships

Quality

63%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./understand-anything-plugin/skills/understand/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.

A highly actionable, well-sequenced multi-phase orchestration skill with exemplary validation feedback loops, but it is verbose and keeps notable content (a full inline validation script, schema tables, normalization lists) in the main body rather than externalizing it as is done for the other bundled scripts.

Suggestions

Move the inline Phase 6 Node.js validation script into a bundled `scripts/validate-graph.cjs` file (consistent with the other externalized scripts) and reference it from the body to cut length and improve progressive disclosure.

Extract the Node/Edge type schema tables and the Phase 4/5 normalization step lists into reference files (e.g. `references/graph-schema.md`, `references/normalization.md`) and link to them from the body.

Trim verbose operational prose — the multi-candidate plugin-root error echo list and repeated issue-number rationale — to tighten conciseness while preserving the decision logic.

Dimension	Reasoning	Score
Conciseness	The body does not pad with concepts Claude already knows (no explanations of git/JSON/knowledge graphs) and is operationally specific, but at ~859 lines it is a dense document with tighten-able prose such as the multi-candidate plugin-root echo list and repeated issue-number rationale, so it is not lean enough for a 3.	2 / 3
Actionability	Guidance is fully executable throughout — complete bash scripts, a full copy-paste Node.js validation script, exact file paths, and concrete JSON shapes with clearly-substituted variables — matching the 'copy-paste ready' anchor.	3 / 3
Workflow Clarity	Seven phases (0–7) are clearly sequenced with explicit validation checkpoints (the Phase 6 validate script, 'fix and re-validate', fingerprint-gate-before-meta.json), feedback loops (retry once on subagent failure), and ordered normalization checklists, so it exceeds the 'clear sequence with validation' anchor.	3 / 3
Progressive Disclosure	References are well-signaled and one level deep (agents/.md, languages/.md, frameworks/.md, locales/.md, bundled scripts), but the body is a monolithic procedure and inlines a full ~60-line validation script that should be an external bundled file like the other scripts, fitting the 'content that should be separate is inline' anchor.	2 / 3
	Total	10 / 12 Passed

Description

50%

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 gives a clear, third-person statement of what the skill does, but it lacks an explicit 'Use when…' trigger clause and broader natural trigger terms, leaving every dimension at the mid-level rather than the top.

Suggestions

Add an explicit 'Use when…' trigger clause (e.g., 'Use when the user wants to understand or visualize a codebase's architecture, or asks for a code map or dependency graph of a project') to answer 'when' and lift completeness.

Broaden trigger-term coverage with natural variations users would actually say — 'understand my code', 'code map', 'dependency graph', 'visualize architecture' — alongside the existing terms.

Name a few concrete sub-actions (scan files, detect languages/frameworks, identify architectural layers, generate a guided tour) to increase specificity beyond the single 'analyze to produce a graph' action.

Dimension	Reasoning	Score
Specificity	It names the domain and one concrete action ('Analyze a codebase to produce an interactive knowledge graph') but does not list multiple specific concrete actions, matching the score-2 anchor 'Names domain and some actions, but not comprehensive' rather than the multi-action score-3 example.	2 / 3
Completeness	It clearly answers 'what' (analyze a codebase to produce a knowledge graph) but has no 'Use when…' clause or equivalent explicit trigger guidance for 'when', so per the judging guidelines completeness is capped at 2.	2 / 3
Trigger Term Quality	Natural terms users would say are present ('codebase', 'architecture', 'understanding', 'components'), but 'interactive knowledge graph' leans technical and common variations like 'code map', 'dependency graph', or 'visualize architecture' are missing, fitting the score-2 'some relevant keywords but missing common variations' anchor.	2 / 3
Distinctiveness Conflict Risk	The 'interactive knowledge graph for understanding architecture' niche is fairly specific, but the generic 'Analyze a codebase' opener could overlap with general code-review/analysis skills and there are no distinct trigger phrases, fitting the score-2 'somewhat specific but could still overlap' anchor.	2 / 3
	Total	8 / 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
skill_md_line_count	SKILL.md is long (859 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: Egonex-AI/Understand-Anything
Path: understand-anything-plugin/skills/understand/SKILL.md
Commit: 6ae7187

Reviewed: 4 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.