he-deepen-plan

Deepen an existing implementation plan so sequencing, verification, and risk treatment are strong enough for execution. Use when the user wants Harness Engineering plan hardening before he-work.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

42%

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

The content is well-structured and concise-leaning with a reasonable procedure and validation note, but its references point to non-existent bundle files and its steps stay abstract rather than providing executable guidance.

Suggestions

Either create the referenced bundle files (contract.yaml, evals.yaml, task-profile.json, routing-map.json, assets) or remove the broken links, since progressive_disclosure is scored against the actual bundle structure.

Make the Procedure actionable by adding a concrete checklist or example gate definitions (e.g. 'each task must list: owner, observable pass/fail criterion, rollback step') instead of generic instructions.

Tighten the Validation and Constraints sections by removing guidance Claude already knows and keeping only skill-specific gates.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and well-organized, but it restates obvious guidance Claude already knows (e.g. 'Ensure each task has observable completion criteria') and some sections add little beyond the description.	2 / 3
Actionability	Steps are concrete directions ('Inspect current plan for ambiguity', 'Deepen task sequencing') but lack executable artifacts like specific commands, checklists, or example gate definitions, leaving them somewhat abstract.	2 / 3
Workflow Clarity	The Procedure lists a 3-step sequence and a 'Fail fast: stop at first failed gate' checkpoint, but validation gates are described generically rather than as explicit validate-fix-retry feedback loops.	2 / 3
Progressive Disclosure	The body points to many external paths (./Infrastructure/references/, ./references, ./assets, ../../../references/), but none of these bundle files or directories exist in the skill bundle, so the references are not real and navigation cannot resolve.	1 / 3
	Total	7 / 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 complete with both capability and an explicit use-trigger, and is reasonably specific to a niche stage, but its trigger language leans on internal jargon rather than natural user phrasing and lists only a couple of concrete actions.

Suggestions

Replace jargon like 'Harness Engineering plan hardening' and 'he-work' with natural user phrasing such as 'tighten an implementation plan before coding' so trigger terms match what users actually say.

Add 1-2 more concrete verbs (e.g. 'assign owners and gates to risky steps, define rollback order, and surface hidden dependencies') to reach the score-3 specificity anchor.

Dimension	Reasoning	Score
Specificity	The description names the domain (plan hardening) and a couple of concrete actions ('Deepen an existing implementation plan', 'sequencing, verification, and risk treatment'), but does not enumerate multiple distinct concrete operations the way the score-3 anchor requires.	2 / 3
Completeness	It answers both 'what' ('Deepen an existing implementation plan so sequencing, verification, and risk treatment are strong enough for execution') and 'when' with an explicit 'Use when the user wants Harness Engineering plan hardening before he-work' clause.	3 / 3
Trigger Term Quality	It surfaces some relevant terms ('implementation plan', 'sequencing', 'verification', 'risk treatment') but relies on jargon ('Harness Engineering plan hardening', 'he-work') rather than natural phrases a user would actually say.	2 / 3
Distinctiveness Conflict Risk	The skill is tied to a niche stage name ('he-deepen-plan', 'he-work') which distinguishes it from generic skills, but the description's framing could still overlap with general planning or review skills when trigger terms are generic.	2 / 3
	Total	9 / 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
metadata_version	'metadata.version' is missing	Warning
relative_links	Relative link issues: 6 missing, 3 deeper-than-1-level, 4 suspicious	Warning

	Total	14 / 16 Passed

Repository: jscraik/Agent-Skills
Commit: 5a6027f

Reviewed: 5 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.