CtrlK
BlogDocsLog inGet started
Tessl Logo

he-improve

Improve existing Harness Engineering skills, references, contracts, and evals from concrete evidence such as failed evals, repeated review findings, usage traces, or documented regressions. Use when a bounded hardening pass is required; do not use for speculative redesign.

59

Quality

67%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./Plugins/harness-engineering/skills/he-improve/SKILL.md
SKILL.md
Quality
Evals
Security

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 well-organized with strong validation and progressive-disclosure structure, but its procedural guidance is more directive than executable and a few dangling reference lines and a duplicate step number detract slightly.

Suggestions

Add at least one concrete, copy-pasteable command or example in the Procedure or Validation section rather than deferring all executable detail to references.

Fix the duplicate '4.' step numbering in the Procedure and convert the trailing '- See references/...' fragments into proper inline links or remove them.

Resolve or relativize the external references that point outside the bundle (e.g., stage-arc-boundary-contract.md) so navigation stays one level deep and verifiable.

DimensionReasoningScore

Conciseness

Lean section/bullet prose that assumes Claude's competence and avoids over-explaining known concepts; a few sentences trail off into dangling reference pointers that could be trimmed.

4 / 5

Actionability

Guidance is specific and directive (e.g., 'Patch one failure class at a time,' 'Apply the git staging contract') but stays at an abstract level with no executable commands or concrete examples in the body itself.

3 / 5

Workflow Clarity

A clear numbered Procedure is paired with a strong fail-fast Validation section, explicit pass|fail|blocked gates, and rollback guidance; minor gaps include a duplicated step number ('4.' appears twice) in the procedure.

4 / 5

Progressive Disclosure

Well-structured overview with clearly signaled one-level references to real bundle files (hot-path-folded-context.md, contract.yaml, evals.yaml all verified present); a couple of external references point outside the bundle, slightly muddying navigation.

4 / 5

Total

15

/

20

Passed

Description

67%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 specific and complete with explicit what/when guidance and a helpful negative-trigger boundary, but its trigger terms skew toward internal jargon rather than natural user phrasing.

Suggestions

Add more natural user-facing trigger phrases (e.g., 'when an eval is failing and the skill needs a bounded fix') alongside the internal terms.

Tighten the core action verb beyond generic 'improve' to a concrete behavior like 'patch and validate'.

DimensionReasoningScore

Specificity

Names the domain (HE skills, references, contracts, evals) and several concrete evidence sources (failed evals, review findings, usage traces, regressions), but the core verb "improve/harden" is fairly generic, leaving minor coverage gaps.

4 / 5

Completeness

Clearly states what it does and includes an explicit "Use when a bounded hardening pass is required" trigger, plus a negative "do not use for speculative redesign" boundary, though the trigger could be more concrete.

4 / 5

Trigger Term Quality

Relevant terms like "failed evals," "review findings," and "regressions" appear, but the phrasing is jargon-heavy and misses the more natural phrases a user would actually say when requesting this skill.

3 / 5

Distinctiveness Conflict Risk

Occupies a clear niche (hardening existing HE surfaces from concrete evidence) with an explicit exclusion reducing conflict risk; minor overlap remains with closely related fix/hardening skills.

4 / 5

Total

15

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

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.