SkillOpt Sleep Cycle

What it does

When your agent uses it

• Letting the agent learn your preferences from past sessions
• Running a one-off offline self-improvement review
• Scheduling a nightly sleep cycle for a project
• Previewing proposed skill edits before applying them

Example prompts

• “Run a dry run of the sleep cycle and show me what it would learn.”
• “Schedule the nightly skillopt cycle for this project.”
• “Adopt the latest staged proposal after backing up my current files.”

Requirements

• The dsh-skillopt plugin with its skillopt tools
• API budget when using a real backend

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

1. 1Harvest — read-only scan of supported local session records → digests
2. 2Mine — digests → recurring task records (intent + outcome labels + checkable refs)
3. 3Replay — re-run tasks under the current skill+memory with the selected backend → (hard, soft) scores
4. 4Consolidate — reflect on failures → propose bounded edits → validation gate on a held-out slice (default: accept only on strict improvement)
5. 5Stage — write accepted proposals to /.skillopt-sleep/staging//. Live files are unchanged. A rejected run still has a report but no…
6. 6Adopt — explicit (or operator-configured --auto-adopt) copies staged files over live ones, backing up first.

What it can do on your machine

Read from SKILL.md and the folder at commit 343db22. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • pip
  • python

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use pip, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

SkillOpt Sleep Cycle loads about 1.4k tokens when it runs. Until then it costs about 146 tokens; SKILL.md has 486 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~146

When it runs · the whole SKILL.md, loaded when a task matches

~1.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from microsoft/SkillOpt at commit 343db22, republished under its MIT licence (© microsoft). 486 words, ~1,436 tokens.

name

skillopt-sleep

description

Use when the user wants the dsh agent to self-improve from past usage, asks about a nightly/offline 'sleep' or 'dream' cycle, skill/memory 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 through the skillopt_* tools: harvest past sessions -> mine recurring tasks -> replay via a selected backend -> consolidate validated skills behind a held-out gate.

# 1. check state (default mock backend, zero cost)
skillopt_status

# 2. preview the cycle
skillopt_dry_run project=<dir> source=<claude|codex|…>

# 3. real run (consumes the selected backend's API budget)
skillopt_run project=<dir> backend=<codex|claude|…> preferences="Prefer pytest; keep commits imperative."

# 4. review the report, then adopt
skillopt_adopt project=<dir>

# 5. schedule nightly at 03:17
skillopt_schedule project=<dir> hour=3 minute=17 backend=<codex>

- insert:
    - id: skillopt
      name: './src/index.js'
      config:
        backend: codex
        project: /path/to/project
        preferences: 'Always use async/await'
        # auto-adopt is OPERATOR-ONLY — the model cannot set it
        autoAdopt: false

pip install skillopt
python -m skillopt_sleep.experiments.run_experiment --persona researcher --assert-improves

© microsoft, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in plugins/dsh/skills/skillopt-sleep of microsoft/SkillOpt.

Open the folder on GitHubat commit 343db22