skillopt-sleep

Use when the user wants their Claude agent to self-improve from past usage, asks about a nightly/offline 'sleep' or 'dream' cycle, memory/skill consolidation, or says things like 'make my agent better the more I use it', 'review my past sessions', 'learn my preferences', 'consolidate what you learned', 'run the sleep cycle', or wants to schedule background self-optimization. Drives the skillopt_sleep engine: harvest past sessions -> mine recurring tasks -> replay through a selected backend -> consolidate validated CLAUDE.md/SKILL.md behind a held-out gate.

Quality

94%

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

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 well-structured, actionable body for a complex skill: concrete commands, a clearly sequenced six-stage cycle with validation gates, and strong safety rules for the destructive adopt step. The main slack is minor over-framing in the intro and a couple of inline sections that could be split into reference files.

Suggestions

Trim the 'synthesizes three ideas' (SkillOpt/Claude Dreams/Agent sleep) and 'deployment-time analogue of training' framing — Claude does not need this conceptual scaffolding to run the skill.

Consider moving the full CLI flag table and the config-keys section into a reference file (e.g. references/cli.md) referenced one level deep, leaving the body with the essential commands and the cycle.

The intro paragraph partly re-states the description's what/when; tighten it to avoid duplicating the frontmatter.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and dense with executable commands, but the opening 'synthesizes three ideas' block and the 'deployment-time analogue of training' framing add context Claude does not strictly need; these could be trimmed.	4 / 5
Actionability	Fully executable, copy-paste-ready commands throughout — the sleep.sh invocations, schedule examples, the CLI flag table, and the python experiment command — covering the common cases concretely.	5 / 5
Workflow Clarity	Six stages are clearly sequenced with explicit validation checkpoints: the held-out gate ('accept only if it strictly improves'), 'Nothing live changes' staging, backup-before-adopt, and 'Evidence before adoption' rule, plus rejected-edit reporting as an error-recovery trail.	5 / 5
Progressive Disclosure	Well-organized into clear sections with signaled one-level-deep references (the bundled sleep.sh script and the external doc URL); the full CLI flag table and config-keys section could arguably live in a separate reference file, but inline placement is reasonable.	4 / 5
	Total	18 / 20 Passed

Description

100%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 strong, third-person description that explicitly pairs a concrete what (the harvest→mine→replay→consolidate pipeline) with an explicit when (named user phrasings and scheduling intent). It is comprehensive and distinguishable, with no over-claims or fluff.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions in sequence — 'harvest past sessions -> mine recurring tasks -> replay through a selected backend -> consolidate validated CLAUDE.md/SKILL.md behind a held-out gate' — giving comprehensive coverage of what the engine does, not just naming the domain.	5 / 5
Completeness	Explicitly answers both 'what' ('Drives the skillopt_sleep engine: harvest... consolidate...') and 'when' via a clear 'Use when the user wants...' clause with concrete trigger phrases, matching the 5-anchor example.	5 / 5
Trigger Term Quality	Comprehensive natural-language triggers with synonyms and variations a user would actually say: 'make my agent better the more I use it', 'review my past sessions', 'learn my preferences', 'consolidate what you learned', 'run the sleep cycle', plus 'sleep'/'dream' cycle phrasings.	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (offline sleep/dream self-improvement cycle) with distinct triggers unlikely to fire for unrelated skills; minimal overlap risk.	5 / 5
	Total	20 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: microsoft/SkillOpt
Path: plugins/claude-code/skills/skillopt-sleep/SKILL.md
Commit: 9c776fc

Reviewed: 3 days 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.