Tessl • The package manager for agent skills

Blog Docs Log inGet started

SkillPublic

code-graph

AST-based code graph for fast symbol lookup, dependency analysis, and blast radius via codebase-memory-mcp MCP server

Quality

69%

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 ./skills/code-graph/SKILL.md

Quality

Content

67%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 well-structured, highly actionable reference for the MCP toolset with a clear workflow and verification step. Its main weaknesses are redundancy — the same graph-vs-grep guidance appears in four forms — and the absence of any progressive disclosure into reference files.

Suggestions

Collapse the 'Decision Framework' and 'Anti-Patterns' sections into the 'When to Use Graph vs Direct Read' table — they restate the same guidance a third and fourth time.

Remove the decorative ASCII box or reduce it to one line; the marketing stats ('64 languages', '~99% fewer tokens') pad tokens without guiding behavior.

Add an error-recovery path to the workflow (e.g., what to do when detect_changes shows a large blast radius or index_status reports a stale graph).

Dimension	Reasoning	Score
Conciseness	The graph-first principle is restated four times across 'Core Principle', the ASCII box, 'Decision Framework', and 'Anti-Patterns', and the box includes marketing padding ('14 MCP tools │ 64 languages │ sub-ms queries │ zero deps', '~99% fewer tokens'). The tool tables are efficient, but the repeated framing and decorative ASCII box mean the body could be meaningfully tightened.	3 / 5
Actionability	All 14 MCP tools are named with purpose, trigger examples ('Who calls sendEmail?', 'Find TODO comments'), a concrete 7-step workflow, and a complete .mcp.json config block with install path. Falls short of 5 only because no actual tool-call syntax or invocation examples are shown.	4 / 5
Workflow Clarity	The PLAN→LOCATE→UNDERSTAND→BLAST→TRACE→CHANGE→VERIFY sequence is clear with an explicit verify step and 'Never skip step 3' emphasis, but there is no error-recovery feedback loop for failure cases (large blast radius, stale graph). Not 3 because validation checkpoints are present and explicit.	4 / 5
Progressive Disclosure	Sections are well-organized with scannable tables and no nested references, and no bundle files exist so nothing is mis-filed. Not 5 because the ~200-line body inlines the full 14-tool catalog and config details that could be split into reference files, leaving no overview-to-detail separation.	4 / 5
	Total	15 / 20 Passed

Description

58%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 concrete, third-person, and names a distinct niche anchored to a specific MCP server. Its main gaps are the missing 'use when' trigger clause and the absence of natural-language phrasings a user would actually say when needing code navigation.

Suggestions

Append an explicit trigger clause, e.g., 'Use when finding function or class definitions, tracing callers, or assessing the impact of a change — before grepping or reading files.'

Add natural synonyms users would say: 'find where a function is defined', 'who calls this', 'callers', 'code search', 'impact analysis'.

Consider mentioning call tracing and architecture overview so the stated capabilities match what the skill actually covers.

Dimension	Reasoning	Score
Specificity	'fast symbol lookup, dependency analysis, and blast radius' lists three concrete capabilities with the mechanism ('AST-based code graph') and vehicle ('via codebase-memory-mcp MCP server'). Not 5 because the body reveals further capabilities (call tracing, architecture overview, code search) that the description omits.	4 / 5
Completeness	The 'what' is clear and concrete, but the description field contains no 'Use when...' clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines. Not 4 because the 'when' is entirely absent from the description itself, not merely implicit.	3 / 5
Trigger Term Quality	'symbol lookup', 'dependency analysis', 'blast radius', and 'code graph' are relevant but technical terms; natural user phrasings like 'find where this function is defined', 'who calls this', or 'impact of a change' are missing. Not 4 because the coverage leans jargon-heavy rather than user-natural.	3 / 5
Distinctiveness Conflict Risk	Naming the specific MCP server and 'AST-based code graph' carves a distinct niche unlikely to trigger for unrelated skills. Not 5 because it could still overlap with generic code-search/navigation skills given no sharper trigger phrasing.	4 / 5
	Total	14 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: alinaqi/maggy
Path: skills/code-graph/SKILL.md
Revision: 55aabe2

Reviewed: 26 days ago

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.

code-graph

Table of Contents

Is this your skill?