skillopt-sleep

Use when the user wants Codex to self-improve from past usage, asks about a nightly/offline 'sleep' or 'dream' cycle, wants Codex to review past sessions, learn preferences, consolidate memory/skills, run dry-run/run/adopt/status for SkillOpt-Sleep, or schedule background self-optimization. Drives the skillopt_sleep engine: harvest past sessions -> mine recurring tasks -> replay through a selected backend -> consolidate validated memory + skills behind a held-out gate.

Quality

85%

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

Quality

Content

81%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 strong, highly actionable body: complete command examples, a well-sequenced cycle with explicit validation and gating, and disciplined token usage. Remaining room is minor — deduplicate the mock-backend note, document the referenced-but-unshown flags (`--tasks-file`, `--legacy`, `--skill`), and consider offloading Windows/config reference material to a reference file.

Suggestions

Document the flags referenced only in passing (`--tasks-file`, `--legacy`, `--skill NAME`/`--all-skills`, `--codex-home`) in the Additional flags table or a usage example, since actionability currently loses points to these undocumented options.

Remove the duplicated mock-backend description (stated both under Actions and under All backends) to tighten conciseness.

Move the Windows launcher variants and the full config-key reference into a small references/ file to reduce the main SKILL.md footprint and improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body is dense and skill-specific — flags, config keys, backends, and safety rules — with no explanation of concepts Claude already knows. Minor duplication (the mock backend's determinism/no-spend is stated twice) and the gbrain experiment details in the Validate section could be trimmed, keeping it below the lean anchor 5.	4 / 5
Actionability	Copy-paste-ready shell commands with flags cover status/harvest/dry-run/run/adopt plus scheduling, Windows variants, and config-key and flag tables. Minor gaps: `--legacy`, `--tasks-file`, `--skill`/`--all-skills`, and `--codex-home` are referenced but never shown in examples or the flag table, so it falls short of anchor 5's fully-executable coverage.	4 / 5
Workflow Clarity	"The cycle" gives a clear 7-step sequence and "Steps" adds an operating procedure with explicit validation checkpoints ("Keep `dry-run --backend mock` as the first smoke check", "read `report.md` before summarizing", "Show validation evidence before recommending adoption") and a stage->review->adopt safety boundary with backups for a batch/destructive operation, matching the anchor-5 pattern of sequenced steps plus explicit validation.	5 / 5
Progressive Disclosure	The body is well-sectioned with clear headers and one clearly signaled external link (RESULTS.md), and no bundle files exist to organize. Minor gaps: the Windows launcher variants, the full backend list, and the config-key reference could live in a reference file to slim the main skill; since this is a ~185-line single-file skill, it is good but not the anchor-5 ideal split.	4 / 5
	Total	17 / 20 Passed

Description

83%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 description: it explicitly pairs a concrete what (the harvest->mine->replay->consolidate engine pipeline) with an explicit, multi-trigger when clause in third-person voice. The only weaknesses are minor — a few natural trigger variants are absent and some pipeline terminology ("held-out gate") is slightly opaque.

Dimension	Reasoning	Score
Specificity	The description enumerates a concrete pipeline ("harvest past sessions -> mine recurring tasks -> replay through a selected backend -> consolidate validated memory + skills") plus named actions (dry-run/run/adopt/status, scheduling), giving several specific actions with only minor gaps; staging/adopt semantics and the phrase "behind a held-out gate" remain slightly under-specified, keeping it below the comprehensive anchor 5.	4 / 5
Completeness	It explicitly answers both questions: an explicit "Use when the user wants…" clause listing concrete trigger scenarios, and a clear "what" describing the engine pipeline. This mirrors the anchor-5 example pattern of concrete what + explicit when.	5 / 5
Trigger Term Quality	It covers natural user phrasings well — "self-improve from past usage", "nightly/offline 'sleep' or 'dream' cycle", "review past sessions", "learn preferences", "consolidate memory/skills", "schedule background self-optimization" — with synonyms (sleep/dream, nightly/offline). A few natural variants (e.g., "learn from past sessions", "get better over time") appear only in the body, not the description, so it falls just short of comprehensive anchor 5.	4 / 5
Distinctiveness Conflict Risk	The niche is distinct (a sleep/self-optimization cycle for the named skillopt_sleep engine, with engine-specific command verbs like dry-run/run/adopt for SkillOpt-Sleep). Generic phrases such as "consolidate memory/skills" or "review past sessions" could overlap with other memory-management skills, so minor overlap risk keeps it below anchor 5.	4 / 5
	Total	17 / 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/codex/skills/skillopt-sleep/SKILL.md
Commit: fa4ca18

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.