dream · wrg32786 — Claude Code skill: lessons and installation
The dream Claude Code skill synthesizes session evidence into lessons and proposed changes. Run it to review recent sessions, consolidate notes, or spot recurring patterns.
MIT ★ 26 updated 2026-11-16
npx skills add wrg32786/aigent-os --skill dream -a claude-codeWhat dream does
The skill gathers evidence from capsules, session logs, failure records, and other sources, then writes lessons and improvement candidates. Lessons go to vault/memory/runtime/LESSONS.jsonl; synthesis goes to vault/memory/DREAM_LOG.md. Each run appends a dated section, even when no lessons qualify.
A lesson is admitted when a pattern recurs across independent events, an operator directive has a separately observed outcome, or a mechanical check supplies exit codes. Generated candidates begin as proposed; only the operator’s exact decision can approve one. The skill does not apply changes or edit system instructions.
An explicit run needs Node and the aigent-os project root. It creates a temporary snapshot, then checks the written result with daemons/nightly-contracts.mjs. A nightly run uses the snapshot made by nightly-pass begin.
What it helps with
- find recurring patterns across sessions
- store evidence-backed lessons in JSONL
- draft testable system-improvement candidates
- filter one-off observations and unsupported claims
Example prompts for the agent
- Run
/dreamto synthesize lessons from recent sessions. - Use
/dream --window nightlyfor the nightly synthesis pass. - Set a candidate’s status with
/dream --set-status <candidate-id> approved --operator-evidence "<direct operator reference>".
How to install dream
1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill dream -a claude-code2Manually: put the files in place
~/.claude/skills/dream/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/dream/
1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill dream -a codex2Manually: put the files in place
.agents/skills/dream/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill dream -a cursor2Manually: put the files in place
.cursor/skills/dream/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install wrg32786/aigent-os skills/dream2Manually: put the files in place
.github/skills/dream/SKILL.md1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill dream -a gemini-cli2Platform CLI
gemini skills install https://github.com/wrg32786/aigent-os --path skills/dreamManually: put the files in place
.devin/skills/dream/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill dream -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about dream
How do I install dream in Claude Code?
Run npx skills add wrg32786/aigent-os --skill dream -a claude-code. The skill file is placed at ~/.claude/skills/dream/SKILL.md.
What does dream do?
It reviews operating evidence, records lessons, and drafts system-improvement candidates. Candidates remain proposed until the operator makes an explicit decision.
What does dream need to run?
An explicit run needs Node and the aigent-os project root. Nightly runs use a snapshot created by nightly-pass begin.
Source text
SKILL.md · MIT
---
name: dream
agent: none
description: "Synthesize lessons and propose improvements from recent operating evidence. Use when reviewing sessions, consolidating notes, or finding patterns."
allowed-tools: Read, Write, Bash
user-invocable: true
status: PRODUCTION - canonical writer and propose-only
triggers:
- dream
- offline consolidation
- consolidation pass
- improvement candidates
- review recent sessions
---
# /dream - evidence-gated synthesis
`/dream` is the canonical writer for lessons and Dream synthesis. It never
applies an improvement, promotes its own proposal, edits system instructions,
or treats an agent-authored claim as verification.
## Canonical ownership
- Lessons: `vault/memory/runtime/LESSONS.jsonl`.
- Dream synthesis: `vault/memory/DREAM_LOG.md`.
- The documented `memory/` fallback is valid only for minimal installations.
- Any older `runtime/DREAM_LOG.md` copy is not an input or output.
- `/close` and `/nightly` invoke this protocol instead of carrying a competing
inline writer.
- Candidate status transitions also go through this writer.
## Invocation
```text
/dream
/dream --window nightly
/dream --set-status <candidate-id> <approved|merged|rejected|blocked> \
--operator-evidence "<direct operator reference>"
```
`proposed -> approved` always requires the operator's exact decision. Never
infer approval from silence, an automated status, another agent, or prior prose.
## Sources
Read, in order:
1. The newest five valid capsules by frontmatter `created_at`.
2. The newest five dated live Session Log blocks.
3. Open Trust Decay captures.
4. The newest three dated Honesty Ledger sessions.
5. Recent Failed Experiments and Failure Modes entries.
6. Existing Lessons JSONL for dedupe and prior evidence chains.
7. Staged `NIGHTLY_CAPTURE_CANDIDATES.jsonl` rows from the prior fire as
discovery and pressure signals only.
8. Explicitly supplied local Git or command evidence.
Record exact source references. A capsule or Session Log entry can establish
that a claim occurred; it cannot prove the claim true.
## Lesson admission predicate
Append a lesson only when one is true:
- the same pattern appears in at least two independent source events;
- one explicit operator directive is paired with a separately observed
outcome; or
- one mechanical red-before/green-after predicate supplies an artifact and
literal exit codes.
Two retellings of one event are not independent. Discard one-off observations,
unverified causal stories, and self-confirmation.
Each admitted lesson has:
- `lesson_id`: next max numeric `lNNN` suffix plus one;
- `type`: `procedural | calibration | strategic | failure-mode | doctrine`;
- `confidence`: `0.0..1.0`, capped at `0.85` for one mechanical event;
- `source_events`: at least two refs unless an allowed exception applies;
- `created_at`: the configured local `YYYY-MM-DD`;
- `admission`: one of:
- `{kind: "two-source"}`;
- `{kind: "operator-outcome", operator_ref, outcome_ref}`;
- `{kind: "mechanical-red-green", red_exit, green_exit: 0, artifact_ref}`.
Serialize UTF-8 JSON, parse the complete JSONL file after append, and confirm
the new line round-trips. Historical duplicate ids do not authorize a new
collision.
## Improvement candidate predicate
When evidence names a repeatable system gap, append this inside the same dated
Dream section:
```markdown
### ni-YYYYMMDD-NN - short name
- **status:** proposed
- **evidence:** source-ref-1; source-ref-2
- **failure class:** concise class
- **proposed_change:** smallest testable change
- **mutation proof:** break condition -> named red; restore -> named green
- **operator gate:** approve / reject / revise
- **operator evidence:** pending
```
Generated candidates always start with the literal field `status: proposed`.
Any generated `approved`,
`merged`, or self-ratifying wording is a hard failure.
## Canonical dated entry
Append one `## YYYY-MM-DD (...)` section every time synthesis runs, even with
zero admitted lessons. It must contain nonempty forms of all eight labels:
- Sources reviewed
- Lessons extracted
- Improvement candidates proposed
- Patterns observed
- Discarded as one-off
- Calibration miscalls
- Productive surprises
- Open trust-decay items to watch
A quiet run writes a minimal zero-lesson entry and appends no Lessons row.
Never overwrite an earlier entry.
## Executable postcondition
A nightly fire uses the snapshot created by `nightly-pass begin`. Explicit
`/dream` creates an operating-system temp snapshot first:
```text
node daemons/nightly-contracts.mjs snapshot --root <aigent-root> \
--out <os-temp>/dream-contract-before.json
```
After writing:
```text
node daemons/nightly-contracts.mjs dream --root <aigent-root> \
--date <YYYY-MM-DD> --snapshot <os-temp>/dream-contract-before.json
```
Only `DREAM_CONTRACT PASS` is green. The validator proves byte-prefix
preservation, parses JSONL, derives ids, validates admission, masks fenced
examples, requires the eight sections and current configured date, and rejects
a newly generated candidate that does not start `proposed`.
Return:
```text
DREAM PASS date=<YYYY-MM-DD> sources=<N> lessons=<N> candidates=<N>
```
© wrg32786 · MIT · wrg32786/aigent-os
Similar skills
SKILL.md skills catalog →dream by other authors comes first.