CtrlK
BlogDocsLog inGet started
Tessl Logo

he-refine

Refine Harness Engineering artifacts, plans, specs, or work into clearer action plans. Use when users ask for tightening, simplification, or lifecycle repair.

56

Quality

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Quality

Content

62%

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-sequenced with strong validation gates, but it is held back by redundant sections, incomplete executable detail, and broken or missing reference links that undermine progressive disclosure.

Suggestions

Add a short, concrete way to resolve and probe the dev server (e.g. a launch-config field to read or a port-probe command) so the procedure is actionable rather than directional.

Link to the actual reference files (contract.yaml, evals.yaml, task-profile.json) from the body and either create the referenced ./assets/ icons or remove the broken asset links.

Collapse the overlapping Philosophy/When to use/Validation/Constraints/Anti-patterns sections to remove restated guidance and improve token efficiency.

DimensionReasoningScore

Conciseness

The body is lean and does not explain concepts Claude already knows, but Philosophy, When to use, Validation, Constraints, and Anti-patterns restate overlapping ideas, so it is mostly efficient but could be tightened.

2 / 3

Actionability

The procedure gives concrete gates ('block if the current branch is a protected default branch', 'one user-noted issue, one focused fix, one quick re-check') but lacks executable commands or specifics for dev-server detection and probing, leaving guidance incomplete.

2 / 3

Workflow Clarity

A clear five-step procedure is paired with explicit validation gates and a 'Fail fast: stop at first failed gate' feedback loop, matching the score-3 anchor for sequenced work with checkpoints.

3 / 3

Progressive Disclosure

Sections are well-organized and the body is concise, but it never links to the real reference files present in the bundle (contract.yaml, evals.yaml, task-profile.json) and its only file links point to a non-existent ./assets/ directory, so references are present but not correctly signaled.

2 / 3

Total

9

/

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 answers both what and when with an explicit trigger, but it names only a narrow set of actions and leans on internal terminology that limits natural trigger coverage and distinctiveness.

Suggestions

List a few more concrete refinement actions (e.g. sharpen acceptance criteria, consolidate duplicate steps, repair stale lifecycles) to lift specificity from 2 to 3.

Add common natural trigger phrases users would actually say (e.g. 'polish', 'clean up', 'sharpen', 'clarify') alongside 'tightening' and 'simplification'.

Tighten the niche so it cannot fire for general refactoring or planning work, e.g. by binding triggers specifically to Harness Engineering artifacts.

DimensionReasoningScore

Specificity

Names the domain ('Harness Engineering artifacts, plans, specs, or work') and a concrete action ('Refine ... into clearer action plans'), but does not list multiple distinct concrete actions like the score-3 anchor, so it stops at naming domain and some actions.

2 / 3

Completeness

Clearly answers both what ('Refine ... into clearer action plans') and when via an explicit 'Use when users ask for tightening, simplification, or lifecycle repair' trigger, matching the score-3 anchor.

3 / 3

Trigger Term Quality

Includes some natural user terms ('tightening, simplification, or lifecycle repair') but misses common variations (polish, clean up, sharpen, clarify) and leans on internal jargon ('Harness Engineering'), so coverage is partial rather than broad.

2 / 3

Distinctiveness Conflict Risk

The 'Harness Engineering' framing gives it a niche, but 'refine plans/specs' and 'simplification' could overlap with general planning or refactoring skills, so it is only somewhat distinguishable rather than a clear non-conflicting niche.

2 / 3

Total

9

/

12

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

relative_links

Relative link issues: 2 missing

Warning

Total

14

/

16

Passed

Repository
jscraik/Agent-Skills
Reviewed

Table of Contents

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.