he-code-review

Review Harness Engineering diffs, PRs, commits, and readiness claims for introduced risk. Use when correctness, validation proof, security posture, traceability, closure safety, or review-thread resolution must be assessed before merge or handoff.

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./Plugins/harness-engineering/skills/he-code-review/SKILL.md

Quality

Content

63%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 content is a well-structured, actionable review doctrine with a sequenced procedure and explicit fail-fast validation. Its weakest points are redundancy around the green-CI/readiness theme and a mismatch between the cited reference paths and the actual bundle contents.

Suggestions

Consolidate the repeated green-CI/readiness warning into one place (e.g. Gotchas) and have the Procedure and Anti-Patterns reference it, to remove triplicate wording.

Reconcile the References block with the actual bundle: either point the "Read when" entries at the real `./references/` files (contract.yaml, evals.yaml, task-profile.json) or move the cited plugin/infrastructure contracts into the bundle.

Trim the Outputs field catalogue to the essential required fields and move the full schema into a reference file, keeping the body as an overview.

Dimension	Reasoning	Score
Conciseness	The body largely assumes Claude's competence (no basic-concept explanations) but repeats the green-CI/readiness warning across Procedure step 7, Gotchas, and Anti-Patterns, and the Outputs section catalogues a long list of required fields, so it could be tightened.	3 / 5
Actionability	It gives concrete directives — exact mode list, real reference paths ("load `../../references/skills/he-code-review/review-mode-contract.md`"), required output fields, and "severity-ranked `file:line` findings" — with only the deepest executable detail deferred to reference files.	4 / 5
Workflow Clarity	A clearly sequenced 13-step Procedure pairs with explicit fail-fast validation ("stop at the first failed gate"), re-verification before accepting, and a blocker-with-recovery-step loop; not a 5 because checkpoints are distributed across separate sections rather than inline at each step.	4 / 5
Progressive Disclosure	Sectioning is clean and the References block signals "Read when" conditions one level deep, but against the actual bundle the cited paths (e.g. `Plugins/harness-engineering/references/...`) do not exist in `./references/`, and the real bundle files (`contract.yaml`, `evals.yaml`, `task-profile.json`) are never referenced from the body.	3 / 5
	Total	14 / 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.

The description cleanly answers both what and when with domain-specific triggers and good natural keyword coverage. Its main limitation is a narrow action vocabulary dominated by "review/assess" rather than multiple distinct concrete actions.

Dimension	Reasoning	Score
Specificity	It names concrete domain objects ("diffs, PRs, commits, and readiness claims") and a concrete purpose ("introduced risk"), but the action vocabulary is essentially a single verb ("Review"/"assess") rather than the several distinct actions the anchor-4 example lists (extracts, fills, converts).	3 / 5
Completeness	Both halves are explicit: the "what" ("Review Harness Engineering diffs, PRs, commits, and readiness claims for introduced risk") and a concrete "Use when ... must be assessed before merge or handoff" trigger clause, matching the anchor-5 example.	5 / 5
Trigger Term Quality	Natural developer terms appear ("PRs", "diffs", "commits", "merge", "handoff", "review-thread"), giving good coverage; not a 5 because synonyms like "pull request" or "code review" are missing and no extension-style triggers apply.	4 / 5
Distinctiveness Conflict Risk	The Harness Engineering niche plus distinctive triggers ("readiness claims", "traceability", "closure safety") make it mostly distinct; not a 5 because "Review diffs, PRs, commits" overlaps with a generic code-review skill.	4 / 5
	Total	16 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: jscraik/Agent-Skills
Path: Plugins/harness-engineering/skills/he-code-review/SKILL.md
Commit: f1f2f21

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.