he-reinforce

Create or refresh evidence-bound Harness Engineering learning artifacts from verified solved problems. Use when a fix worked, a repeated failure should become durable knowledge, or .harness/solutions and Project Brain need maintenance.

Quality

68%

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 ./Plugins/harness-engineering/skills/he-reinforce/SKILL.md

Quality

Content

62%Weight 40%Scale 1-3

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

A well-structured, deliberately-disclosed team-automation skill with strong explicit validation and failure-handling feedback loops. It is held back by repeated identical reference pointers, several truncated/broken sentences, abstract procedure steps, and a cluttered reference list mixing in-bundle and external paths.

Suggestions

Replace the ~8 repeated 'See references/hot-path-folded-context.md for folded X detail.' lines with a single, consolidated pointer near the top of the relevant sections to cut redundant tokens.

Repair the truncated sentences (e.g. '...command boundaries, and approval' and '...a verified repo learning, decision, invariant, or') so each list item is grammatically complete and actionable.

Distinguish in-bundle references from shared '../../references/' paths in the References section (e.g. a 'Shared HE references' subsection) so navigation is unambiguous, and tighten the abstract procedure steps with the concrete gates already named in the Validation section.

Dimension	Reasoning	Score
Conciseness	The body is largely lean and does not explain concepts Claude already knows, but it repeats the identical 'See references/hot-path-folded-context.md for folded X detail.' line ~8 times and contains broken/truncated sentences (e.g. '...command boundaries, and approval', '...learning, decision, invariant, or'), which waste tokens and reduce tightness.	2 / 3
Actionability	Concrete elements exist (named modes like capture_solved_problem, paths like '.harness/solutions/**', and the executable 'python3 Plugins/harness-engineering/scripts/check_bluf_structure.py <path> --json'), but the core procedure steps stay abstract ('Prove eligibility', 'Keep scope tight') and defer most detail to references.	2 / 3
Workflow Clarity	There is a numbered 5-step procedure plus a dedicated Validation section that enumerates gates, mandates pass/fail/blocked reporting, and supplies a concrete validation script; Failure Handling then forms an explicit feedback loop routing blocked work to the right HE stage.	3 / 3
Progressive Disclosure	References are one-level-deep, clearly signaled, and the local bundle paths all resolve to real files, but the same folded-context pointer is repeated redundantly across sections, six '../../references/' paths sit outside this bundle, and the trailing context-disposition policy note clutters the organization.	2 / 3
	Total	9 / 12 Passed

Description

75%Weight 40%Scale 1-3

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, third-person description that clearly answers both what and when with explicit triggers and a distinct niche. Its weakness is jargon-heavy trigger terms and only two named actions, which limit specificity and natural-term coverage.

Suggestions

Add one or two more concrete actions (e.g. 'consolidate', 'redact', 'sync') so the capability list reads as multiple distinct operations rather than just 'create or refresh'.

Soften the trigger clause with at least one plain-language variant (e.g. 'when a fix worked and is worth remembering') alongside the internal terms so non-HE-fluent users can still match it.

Dimension	Reasoning	Score
Specificity	Names the domain ('evidence-bound Harness Engineering learning artifacts from verified solved problems') and the actions 'Create or refresh', but only two verbs are listed rather than multiple distinct concrete actions, so it is not comprehensive.	2 / 3
Completeness	It states what the skill does ('Create or refresh ... learning artifacts') and gives an explicit 'Use when ...' trigger clause, so both what and when are clearly answered.	3 / 3
Trigger Term Quality	The 'Use when a fix worked, a repeated failure should become durable knowledge' clause supplies several natural phrases, but triggers lean heavily on internal jargon ('.harness/solutions', 'Project Brain') and miss common layperson variations.	2 / 3
Distinctiveness Conflict Risk	The Harness Engineering learning-artifact niche with triggers tied to '.harness/solutions' and 'Project Brain' is distinct and unlikely to fire for unrelated skills.	3 / 3
	Total	10 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: jscraik/Agent-Skills
Path: Plugins/harness-engineering/skills/he-reinforce/SKILL.md
Commit: 46e4be2

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.