context-map

Maps file dependencies, flags shared imports, groups files for safe parallel editing before code changes. Use when planning a refactoring, analyzing change impact, or understanding which files a modification will affect.

Quality

83%

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

78%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.

A lean, well-structured skill with genuinely executable search commands and a concrete output template. The main defects are the dangling REFERENCE.md pointer (the bundle contains no reference files), the lack of an error-recovery loop at the validation checkpoint, and example-specific commands without genericized placeholders.

Suggestions

Ship the referenced REFERENCE.md (or remove the pointer and inline the minimal template details) — the link currently resolves to nothing, breaking progressive disclosure entirely.

Add a short feedback loop at the validation checkpoint: what to do when a listed file fails typecheck or a search hit doesn't fit the map (e.g., re-trace that subtree or correct the entry-point list).

Generalize the rg commands with explicit placeholder symbols (e.g., 'rg -n "import .*from .*<lib-path>"' with a note to substitute the entry-point export) so the commands model the pattern, not just one example.

Dimension	Reasoning	Score
Conciseness	The ~70-line body is lean and assumes Claude's competence: no concept explanations, just operational steps, concrete rg commands, an output template, and terse anti-patterns. Nothing beyond a formatting nit (an empty 'Skip' table cell) is trimmable.	5 / 5
Actionability	Concrete, copy-paste-able rg commands for both trace directions ('rg -n "import .from .libs/auth"', 'rg -n "from .*libs/" src/auth/') plus a full markdown output template. Not a 5 because the commands hardcode example symbols (libs/auth, AuthCard, priceRange) that the reader must substitute, with no generic placeholder pattern shown.	4 / 5
Workflow Clarity	Clear 4-step sequence (entry points → trace outward → trace inward → build map) with an explicit validation checkpoint ('every search hit lands in the map; all listed files open without errors'). Not a 5 because there is no error-recovery loop and the typecheck is delegated vaguely to a 'codebase-tool slot' rather than a runnable command; not a 3 because the checkpoint is explicit, not implicit.	4 / 5
Progressive Disclosure	The body correctly keeps detail out of SKILL.md and clearly signals a one-level-deep pointer ('Full template, depth levels, partition derivation, Team Lead integration: see [REFERENCE.md](./REFERENCE.md)'), but that file does not exist in the bundle (no references/ directory at all), so the disclosure promise fails at load time. Scoring against the actual bundle structure per the rubric guideline, this lands at 'some structure but could be better organized' rather than a working split.	3 / 5
	Total	16 / 20 Passed

Description

88%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.

A strong description: concrete third-person actions, an explicit 'Use when' clause with natural trigger phrases, and a distinct niche. Only minor improvements available — broader trigger synonyms and slightly sharper separation from general planning skills.

Dimension	Reasoning	Score
Specificity	Three concrete, distinct actions in the dependency-mapping domain: 'Maps file dependencies, flags shared imports, groups files for safe parallel editing'. Comprehensively covers what the skill does; not a 4 because there are no coverage gaps in the action list.	5 / 5
Completeness	Explicitly answers both halves: a clear 'what' (the three mapped/flagged/grouped actions) and an explicit 'Use when' clause with three concrete trigger scenarios. Third-person voice throughout, so no voice penalty applies.	5 / 5
Trigger Term Quality	Natural user phrases are present: 'planning a refactoring', 'analyzing change impact', 'which files a modification will affect'. Not a 5 because common synonyms are missing (e.g., 'dependency graph', 'affected files', 'blast radius'); not a 3 because coverage goes beyond a few relevant keywords.	4 / 5
Distinctiveness Conflict Risk	Pre-change dependency mapping for safe parallel editing is a clear, well-defined niche. Minor overlap risk with generic refactoring-planning or code-review skills; not a 5 because 'planning a refactoring' could also plausibly invoke those.	4 / 5
	Total	18 / 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
relative_links	Relative link issues: 1 missing	Warning

	Total	15 / 16 Passed

Repository: monkilabs/opencastle
Path: src/orchestrator/skills/context-map/SKILL.md
Commit: c0f4566

Reviewed: 1 day 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.