investigate

Systematic debugging with Iron Law methodology. 5-phase investigation from evidence collection to verified fix. Triggers on "investigate", "debug", "root cause".

Quality

67%

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 ./.claude/skills/investigate/SKILL.md

Quality

Content

56%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 well-sequenced debugging workflow with a clear gate and escalation rule, but it repeats the phase list and offers only a few concrete commands while most phases stay abstract. Referenced bundle files are missing, undermining progressive disclosure.

Suggestions

De-duplicate the 5-phase description: keep the Execution Flow as the canonical sequence and reference it from the Iron Law section instead of restating the numbered list.

Add concrete executable commands for Phases 2-3 (e.g., the actual grep/CLI commands for pattern search and hypothesis probes) so the body is copy-paste ready, not just structured pseudocode.

Create the referenced phases/*.md and specs/*.md files (or remove the Reference Documents table) so the one-level-deep navigation actually resolves.

Dimension	Reasoning	Score
Conciseness	Mostly efficient prose, but the 5 phases are described three times (Iron Law Principle numbered list, Execution Flow ASCII diagram, and Reference Documents table), which is real tightening opportunity; not a 4 because the repetition is noticeable.	3 / 5
Actionability	Provides a couple concrete commands ('mkdir -p .workflow/.debug', 'ccw cli --tool gemini --mode analysis') and an output structure, but most phases are described as abstract steps ('Reproduce bug, collect evidence', 'Form max 3 hypotheses') without executable commands, reading like structured pseudocode.	3 / 5
Workflow Clarity	Clear 5-phase sequence with an explicit GATE in Phase 4, a 3-strike escalation rule, and Phase 5 verification/regression checkpoints; just below 5 because some validation feedback loops (e.g., re-validate after a failed hypothesis test) are implied rather than fully spelled out.	4 / 5
Progressive Disclosure	Sections are well organized and references are signaled in a table, but the Reference Documents table points to phases/.md and specs/.md files that do not exist in any bundle directory, so navigation is effectively broken.	3 / 5
	Total	13 / 20 Passed

Description

78%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 solid description that explicitly states what the skill does and when to invoke it with concrete trigger phrases. Specificity and distinctiveness are good but not maximal due to process-level phrasing and minor overlap with generic debugging.

Dimension	Reasoning	Score
Specificity	Names the debugging domain and concrete actions ('evidence collection', 'verified fix', '5-phase investigation'), but describes process phases rather than comprehensive concrete capabilities; not a level-4 list of several specific actions.	3 / 5
Completeness	Explicitly answers both 'what' (systematic debugging, 5-phase investigation from evidence collection to verified fix) and 'when' (concrete trigger phrases 'investigate', 'debug', 'root cause').	5 / 5
Trigger Term Quality	'Triggers on "investigate", "debug", "root cause"' uses natural terms a user would say, but misses common synonyms like 'troubleshoot' or 'why is X failing', so not a 5.	4 / 5
Distinctiveness Conflict Risk	'Iron Law methodology' and the specific trigger set carve a clear niche with mostly distinct triggers, but 'debug' has minor overlap risk with general coding skills, keeping it just below 5.	4 / 5
	Total	16 / 20 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
relative_links	Relative link issues: 7 missing	Warning

	Total	14 / 16 Passed

Repository: catlog22/Claude-Code-Workflow
Path: .claude/skills/investigate/SKILL.md
Commit: 07491b0

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.