CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

he-strategy

Compress HE cognition artifacts into evidence-backed strategy. Use when intent, review, triage, ADR, core, or source-prompt comparison evidence needs durable direction.

Quality

82%

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

SKILL.md
Quality
Evals
Security

Quality

Content

88%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.

A tightly compressed, well-structured process skill: lean prose, a fully sequenced procedure with explicit validation gates and rerun loops, and condition-based routing to reference files. The main gaps are the absence of any concrete example artifact or command invocation and heavy reliance on out-of-bundle shared references.

DimensionReasoningScore

Conciseness

The body is dense, unpadded directive prose: every section states policy or instructions with no explanation of concepts Claude already knows and no filler (e.g., 'Strategy artifacts are cognition compression, not ceremony.'). Rewarding brevity that preserves clarity, this matches the 'lean and efficient; every token earns its place' anchor.

5 / 5

Actionability

The 10-step Procedure gives concrete, executable process guidance: named paths ('../../references/routing-map.json'), naming rules, and a Validation section with specific gates ('skill_gate.py', 'Vale', 'skill_gate ... progressive-disclosure lint') plus example user asks. It stops short of 5 because there is no concrete sample artifact or command invocation, and several steps remain policy-level ('Compress conclusions to decisions that change routing, deletion, investment...') without showing how.

4 / 5

Workflow Clarity

The Procedure is a clearly sequenced 10-step workflow with explicit validation checkpoints ('record each gate as pass, fail, or blocked'), an error-recovery feedback loop ('Rerun failed gates'), preconditions, and a dedicated Failure Mode section with stop/pivot conditions. This matches the anchor for clear sequence with explicit validation steps and feedback loops.

5 / 5

Progressive Disclosure

The body is an overview that routes mode-specific detail to one-level-deep reference files via a well-signaled References section with 'Read when <condition> -> <path>' routing, and locally referenced files (references/contract.yaml, references/evals.yaml, references/strategy-output-contract.md) exist in the bundle. It is not 5 because most referenced paths ('../../references/...') point outside the bundle where their existence cannot be verified, and the body still inlines considerable policy detail (Constraints, Gotchas, Anti-Patterns, Stage Arc Boundary) that could arguably be routed to references.

4 / 5

Total

18

/

20

Passed

Description

70%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.

A concise, niche-specific description with an explicit 'Use when' trigger clause covering the skill's six modes. Its main weakness is abstraction: the single action verb 'Compress ... into evidence-backed strategy' and the fuzzy condition 'needs durable direction' leave the what/when less concrete than top-scoring examples.

Suggestions

Specificity: replace the abstract 'Compress HE cognition artifacts into evidence-backed strategy' with 2-3 concrete actions, e.g., 'Extract repo intent, compress review/triage evidence into strategy artifacts, and validate ADR/core decisions against source evidence.'

Completeness: make the 'when' triggers concrete user-facing phrases instead of 'needs durable direction', e.g., 'Use when the user asks for repo intent, an architecture-review summary, triage direction, an ADR/core decision check, or a source-prompt equivalence comparison.'

Trigger term quality: add natural synonyms users would actually say (e.g., 'strategy summary', 'decision record', 'architecture direction') alongside the internal mode names.

DimensionReasoningScore

Specificity

The description names its domain ('HE cognition artifacts') and one core action ('Compress ... into evidence-backed strategy'), matching the anchor for 1-2 concrete actions without comprehensive coverage. It does not list several specific actions, so it does not reach 4.

3 / 5

Completeness

Both 'what' ('Compress HE cognition artifacts into evidence-backed strategy') and 'when' (an explicit 'Use when intent, review, triage, ADR, core, or source-prompt comparison evidence needs durable direction') are present. It is not 5 because the 'when' condition hinges on the fuzzy phrase 'needs durable direction', and bare words like 'review' and 'intent' lack qualifiers that would make triggers fully concrete.

4 / 5

Trigger Term Quality

The 'Use when' clause covers six natural domain terms (intent, review, triage, ADR, core, source-prompt comparison), giving good keyword coverage a user in this ecosystem would say. It falls short of 5 because common variations/synonyms are missing and 'durable direction' is internal jargon rather than a natural user phrase.

4 / 5

Distinctiveness Conflict Risk

The description carves a clear strategy-compression niche that is distinct from implementation skills, and its trigger terms are domain-specific. Minor overlap risk remains with closely related sibling skills (e.g., he-reframe, he-spec) that share 'review' and 'triage' vocabulary, keeping it below 5.

4 / 5

Total

15

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository

jscraik/Agent-Skills

Path

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

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.