CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

decay

Manage evidence freshness by identifying stale decisions and providing governance actions

Quality

67%

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 ./antigravity/skills/decay/SKILL.md

The canonical home for this skill is decay in NeoLabHQ/context-engineering-kit

SKILL.md
Quality
Evals
Security

Quality

Content

68%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 body is a well-structured, largely actionable operational guide: concrete report formats, file paths, and copy-paste-ready record templates. Its main weaknesses are the absence of validation/verification steps around the destructive deprecate file-moves and batch evidence scan, and a Refresh branch that says what to do without how to do it.

Suggestions

Add explicit validation checkpoints around the destructive Deprecate action — e.g., verify the hypothesis file exists before the `mv`, confirm the move succeeded, and verify the deprecation record was written — to lift workflow clarity above the cap.

Make the Refresh branch concrete: specify which command or procedure re-runs validation and where the fresh evidence file lands, instead of the one-line 'Re-run validation to create fresh evidence'.

Deduplicate the Three Actions table (Quick Concepts vs. Step 3) and consider moving the full deprecation/waiver record templates into a references file to tighten the main body.

DimensionReasoningScore

Conciseness

The body is dominated by tables, file templates, and commands, with the key principle stated once ('Evidence is perishable. Decisions built on expired evidence carry hidden risk.'); only minor instances could be trimmed, such as the Three Actions table being duplicated between Quick Concepts and Step 3 and illustrative prose like 'A benchmark from 6 months ago may no longer reflect current system performance.'

4 / 5

Actionability

Guidance is mostly executable: exact paths (`.fpf/evidence/`, `.fpf/knowledge/L2/`), a concrete `mv` command, and complete record templates with frontmatter for both deprecations and waivers. The Refresh branch is the one gap — 'Re-run validation to create fresh evidence' gives no procedure or pointer.

4 / 5

Workflow Clarity

The report → present → act sequence is clearly laid out with a concrete report format, but the Deprecate action moves knowledge files (destructive) and Step 1 scans all evidence files (batch) with no verification or confirmation checkpoints, which caps workflow clarity at 3 per the rubric guidelines.

3 / 5

Progressive Disclosure

This is a single-file skill with no bundle files, and it is well-sectioned with clear headers and no nested references; the full record templates inlined in Step 3 could arguably live in a references file, which is the only minor organization gap.

4 / 5

Total

15

/

20

Passed

Description

50%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 communicates a clear domain and purpose but stops at a mid-level abstraction: it never names the three concrete governance actions, includes no usage triggers, and leans on the generic verb 'Manage'. Adding a 'Use when...' clause and naming refresh/deprecate/waive would lift most dimensions.

Suggestions

Add an explicit trigger clause, e.g. 'Use when evidence has expired or decisions depend on outdated benchmarks, audits, or test results.'

Replace the vague 'providing governance actions' with the concrete actions: refresh stale evidence, deprecate obsolete decisions, or waive risk with a documented deadline.

Include natural trigger-term variations users would actually say — 'expired evidence', 'outdated decision', 'waiver' — to improve trigger term quality and distinctiveness.

DimensionReasoningScore

Specificity

Names the domain ('evidence freshness') and one concrete action ('identifying stale decisions'), but 'providing governance actions' is unspecified — the actual actions (refresh, deprecate, waive) are absent, so coverage is incomplete.

3 / 5

Completeness

The 'what' is clear (identify stale decisions, provide governance actions), but there is no 'Use when...' clause or equivalent trigger guidance, which caps completeness at 3.

3 / 5

Trigger Term Quality

'evidence freshness' and 'stale decisions' are relevant domain terms, but common natural variations users would say — 'expired', 'outdated', 'refresh', 'waive' — are missing.

3 / 5

Distinctiveness Conflict Risk

'Evidence freshness' and 'stale decisions' carve a niche, but the generic verb 'Manage' and unspecified 'governance actions' leave overlap risk with other evidence- or decision-management skills.

3 / 5

Total

12

/

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

NeoLabHQ/context-engineering-kit

Path

antigravity/skills/decay/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.