CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

sy-work

Build only the approved slice while preserving scope, unrelated worktree changes, and validation evidence. Use when an approved execution plan or explicit bounded task is ready for implementation.

Quality

76%

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/synaipse-harness/skills/sy-work/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

78%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 well-organized, token-efficient stage skill with strong progressive disclosure and a clear validated workflow. Its main weakness is actionability: executable specifics (the stage-result schema, allowed/forbidden writes, actual validation commands) are entirely deferred to references/contract.yaml, leaving the body's instructions somewhat abstract.

Suggestions

Inline one minimal filled-in stage-result example artifact so the Deliverables field list becomes concrete without adding significant tokens.

Either name the actual plugin validation and SDK proof commands (or point to the exact section of contract.yaml that defines them) in the Validation section.

Replace circular phrasing like 'the work owned by work' with the concrete allowed-writes boundary from references/contract.yaml ('approved source files for the slice and local evidence artifacts named by the plan').

DimensionReasoningScore

Conciseness

The ~55-line body is lean and efficient: every section (When to use, Required inputs, Deliverables, Procedure, Validation, Handoff, Failure modes, Gotchas) is short, nothing explains concepts Claude already knows, and no token is wasted on padding.

5 / 5

Actionability

Guidance is sequenced but high-level: 'Perform only the work owned by work' is circular, 'run plugin validation and SDK proof commands from the repository wrapper' names no actual commands, and no example stage-result artifact is shown inline — the concrete schema and write boundaries live in references/contract.yaml. Field names (schema_version, evidence_refs, blocked_by) and status labels (pass, fail, blocked, not_checked) provide some concreteness, keeping this above anchor 2 but below anchor 4.

3 / 5

Workflow Clarity

The 5-step Procedure is clearly sequenced, with a fail-fast validation checkpoint ('stop at the first failed required gate'), an exit-criteria gate on handoff, and failure-mode recovery routing. It stops short of anchor 5 because there is no explicit validate-fix-retry loop within the procedure itself; failure handling routes outward rather than looping back.

4 / 5

Progressive Disclosure

The body is a clear overview with four well-signaled one-level-deep references (contract.yaml, evals.yaml, task-profile.json, source-context.yaml), all of which exist in the bundle and are substantive; no nested reference chains or inlined bulk content.

5 / 5

Total

17

/

20

Passed

Description

62%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 lean and correctly structured with an explicit 'Use when' trigger clause, but it is written in internal Harness jargon ('approved slice', 'evidence lanes') that weakens natural trigger matching and undersells the concrete deliverables the stage produces. It sits comfortably above mediocre but below the exemplary good_overall_examples.

Suggestions

Replace jargon like 'the approved slice' with the concrete output, e.g. 'Emit a stage result (schema_version, stage, status, evidence_refs, next_stage) after implementing the approved execution plan'.

Add natural trigger variations users would actually say, such as 'implement', 'execute the approved plan', or 'work stage', to improve trigger term coverage.

Name the validation classification deliverable (pass/fail/blocked) in the description so the 'what' answers what the skill produces, not just what it guards.

DimensionReasoningScore

Specificity

The description names real actions ('Build only the approved slice while preserving scope, unrelated worktree changes, and validation evidence') but relies on internal jargon ('approved slice', 'scope') and omits the stage's deliverables (stage result, validation classification, next-stage handoff), so coverage is not comprehensive.

3 / 5

Completeness

Both 'what' ('Build only the approved slice while preserving...') and an explicit 'when' ('Use when an approved execution plan or explicit bounded task is ready for implementation') are present; the 'what' is present but not fully concrete about outputs, so it falls short of anchor 5.

4 / 5

Trigger Term Quality

'approved execution plan', 'explicit bounded task', and 'ready for implementation' are relevant trigger terms, but common natural variations users would say ('implement the plan', 'execute the plan', 'work stage') are missing.

3 / 5

Distinctiveness Conflict Risk

The description carves a clear niche (implementing an approved plan/bounded task within a staged lifecycle), with only minor overlap risk against generic implementation skills.

4 / 5

Total

14

/

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.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

jscraik/Agent-Skills

Path

Plugins/synaipse-harness/skills/sy-work/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.