he-work

Executes bounded implementation work from approved specs, plans, issues, or small fixes by editing files, running validation, preserving unrelated work, and recording rollback and handoff evidence. Use when asked to implement, execute a plan, apply changes, or build a scoped feature.

Quality

65%

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-work/SKILL.md

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 body is well-organized with a clear sequenced workflow and explicit validation/feedback gates, scoring well on workflow clarity. It is held back by redundancy across the constraints/gotchas sections, mostly abstract actionability, and a broken reference structure where signaled files are not in the bundle and the available reference is orphaned.

Suggestions

Collapse the overlapping Constraints, Anti-patterns, Gotchas, and Failure mode sections into a single "Rules / stop conditions" list to remove duplicated statements like "Do not claim completion without validation evidence".

Make slice guidance concrete: define what counts as a verified slice and which validation command confirms it, beyond the single lint script.

Fix reference wiring: either link the existing references/context-preservation.md from the body, or point the References section at files that actually exist in this skill's bundle rather than external Plugins/ paths.

Dimension	Reasoning	Score
Conciseness	Sections are individually lean, but Constraints, Anti-patterns, Gotchas, and Failure mode overlap heavily (e.g., "Do not claim completion without validation evidence" and "Do not expand scope" each recur), so the body could be tightened.	2 / 3
Actionability	One concrete executable command is present (he_linear_traceability_lint.py), but most guidance ("Implement one verified slice at a time", "Resolve Linear, branch, PR, IDs") is abstract direction rather than fully executable specifics.	2 / 3
Workflow Clarity	The numbered Procedure pairs with an explicit Validation section and feedback loops (reflect contract drift in the governing artifact before continuing; stop at the first failed gate), matching the clear-sequence-with-validation anchor.	3 / 3
Progressive Disclosure	References are clearly signaled one level deep, but all seven cited paths point to external Plugins/ locations absent from this bundle, and the one real bundle file (references/context-preservation.md) is never linked, so navigation does not resolve cleanly.	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 cleanly covers both what and when with an explicit Use-when trigger, but its action vocabulary is high-level and its triggers mix natural terms with internal jargon. It is distinct enough to signal a niche yet could still overlap with general implementation skills.

Suggestions

Replace the single verb "Implement" with 2-3 concrete actions (e.g., "implement, validate, and ship approved work slices") to lift specificity toward the multi-action anchor.

Add common natural phrasings users would actually say (e.g., "implement this plan", "work through this todo list", "ship this approved spec") alongside the internal terms.

Sharpen distinctiveness by foregrounding the traceability/Linear gating that separates this from generic coding skills, so the trigger cannot fire for ordinary implementation requests.

Dimension	Reasoning	Score
Specificity	"Implement approved Harness Engineering work" names a domain and a single action but does not list multiple specific concrete actions, matching the score-2 anchor rather than the multi-action score-3 example.	2 / 3
Completeness	It explicitly answers both what ("Implement approved Harness Engineering work") and when ("Use when a plan, todo list, or tiny spec needs traceable delivery and validation"), satisfying the explicit-trigger score-3 anchor.	3 / 3
Trigger Term Quality	"plan, todo list, or tiny spec" are reasonably natural triggers, but "Harness Engineering", "traceable delivery", and "validation" lean toward internal jargon rather than broad natural phrasing a user would say.	2 / 3
Distinctiveness Conflict Risk	The "Harness Engineering" / approved-artifact framing gives a niche, but the core trigger "implement a plan/todo list" overlaps broadly with generic implementation skills, so it is not yet a clearly non-conflicting niche.	2 / 3
	Total	9 / 12 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-work/SKILL.md
Commit: 46e4be2

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.