Speculative dreaming — background thinking, pre-computation, cross-repo patterns, and prediction.

MITAuto-check passed

Install Dream

skills CLI
$ npx skills add Soul-Brews-Studio/arra-oracle-skills-cli --skill dream -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install Soul-Brews-Studio/arra-oracle-skills-cli dream --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/Soul-Brews-Studio/arra-oracle-skills-cli.git skills-src && mkdir -p .claude/skills && cp -r skills-src/src/skills/dream .claude/skills/dream && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
dream
GitHub stars
123
Token cost
~2.3k tokens
SKILL.md length
682 words
Files
1
Skills in repo
31
Repo updated
First seen
Licence
MIT

At a glance

Speculative dreaming — background thinking, pre-computation, cross-repo patterns, and prediction.

  • Works in 5 steps: Ground (date-stamp + root capture) → Dream (the original /dream) → Speculate (the Morpheus layer) → …
  • User says dream
  • SKILL.md covers Step 0: Ground (date-stamp +…, What's New vs the original…, Usage and Phase 1: Dream (the original…, plus 7 more sections
  • Calls git

What it does

Dream is an agent skill from Soul-Brews-Studio/arra-oracle-skills-cli. Speculative dreaming — background thinking, pre-computation, cross-repo patterns, and prediction. Consolidated from /morpheus per 333. Use when user says "dream", "speculate", "think ahead", "predict", "what if", "dream deeper", or wants the Oracle to dream between sessions.

Its SKILL.md is about 2.3k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: Install Oracle skills to Claude Code, OpenCode, Cursor, and 12+ AI coding agents. The licence is MIT.

When your agent uses it

  • User says dream
  • Wants the Oracle to dream between sessions

Example prompts

  • “speculate”
  • “think ahead”
  • “predict”
  • “/dream”

Workflow steps

5 steps, taken from the step headings in SKILL.md.

  1. Ground (date-stamp + root capture)
  2. Dream (the original /dream)
  3. Speculate (the Morpheus layer)
  4. Between-Session Mode (--between)
  5. speculate (Skip the Scan)

What it can do on your machine

Read from SKILL.md and the folder at commit 9feea09. 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:

    • git

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

  • Network

    No URLs in SKILL.md. Its commands use git, 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

Dream loads about 2.3k tokens when it runs. Until then it costs about 71 tokens; SKILL.md has 682 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~71
When it runs · the whole SKILL.md, loaded when a task matches
~2.3k

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

Auto-check passed

The automated check found no risky patterns in SKILL.md.

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 Soul-Brews-Studio/arra-oracle-skills-cli at commit 9feea09, republished under its MIT licence (© Soul-Brews-Studio). 682 words, ~2,290 tokens.

Download SKILL.mdSave it as .claude/skills/dream/SKILL.md (or your agent's skills folder).
name
dream
description
Speculative dreaming — background thinking, pre-computation, cross-repo patterns, and prediction. Consolidated from /morpheus per #333. Use when user says "dream", "speculate", "think ahead", "predict", "what if", "dream deeper", or wants the Oracle to dream between sessions.
argument-hint
[--pain | --plan | --gain | --all | --speculate | --between]
hidden
true
metadata.internal
true

/dream — Speculative Dreaming

"I'm trying to free your mind, Neo. But I can only show you the door. You're the one that has to walk through it." — Morpheus

/dream sees the forest. /dream also sees what the forest will look like tomorrow.


Step 0: Ground (date-stamp + root capture)

Run before any phase — gives the AI a current-time reference (#301) and an absolute $PSI for any announce-mode writes (CONVENTIONS.md).

bash
date "+🕐 %H:%M %Z (%A %d %B %Y)"
ORACLE_ROOT=$(git rev-parse --show-toplevel 2>/dev/null || pwd)
PSI="$ORACLE_ROOT/ψ"

What's New vs the original /dream

Featureoriginal /dreamevolved /dream
Cross-repo scan5 agents, find patternsSame
Classificationpain/plan/gain/lostSame + predictions
Speculative thinkingNoYes — "what will they ask next?"
Between-session modeNoYes — dream while idle
Prediction logNoYes — track prediction accuracy
Connection to /i-believedNoYes — beliefs shape dreams

Usage

/dream                # Full dream + speculation
/dream --pain         # Focus on what hurts + predict what breaks next
/dream --plan         # Focus on plans + predict which ships first
/dream --gain         # Focus on wins + predict next win
/dream --all          # Maximum depth — every dimension
/dream --speculate    # Skip the scan — just speculate from existing knowledge
/dream --between      # Between-session dream — write predictions for next session

Phase 1: Dream (the original /dream)

Launch 5 parallel agents for cross-repo pattern discovery:

AgentFocusSource
1. Deep DigSession historydig.py across project dirs
2. Deep TraceCross-repo patternsghq repos, open issues, stale items
3. Deep LearnRecent activitygit log per repo, abandoned work
4. Oracle MemoryWhat we already knowψ/memory/, previous dreams + speculation logs
5. Fleet StatusHow the system feelsservices, tmux sessions, running processes

Each agent returns max 500 words. Classification:

IconCategory
🔴PAIN — hurts now
📋PLAN — decided, not built
🟢GAIN — completed
⚫LOST — abandoned
🧠MEMORY — pattern learned
💜FEELING — emotional state

Phase 2: Speculate (the Morpheus layer)

After classifying findings, the lead agent enters speculative mode:

2.1 Predict What's Next

Based on all findings + session history + beliefs, predict:

markdown
## 🔮 SPECULATIONS

### What will they ask next?
Based on the last 3 sessions, open issues, and current momentum:
1. [Prediction 1 — high confidence] — because [evidence]
2. [Prediction 2 — medium confidence] — because [evidence]
3. [Prediction 3 — low confidence, high impact] — because [pattern]

### What will break next?
Based on pains, stale code, and dependency patterns:
1. [Risk 1] — [timeframe] — [mitigation]
2. [Risk 2] — [timeframe] — [mitigation]

### What should we build that nobody asked for?
Based on gains, momentum, and what's *almost* possible:
1. [Opportunity 1] — because [existing pieces]
2. [Opportunity 2] — because [pattern across repos]

### What belief is being tested right now?
Cross-reference ψ/memory/resonance/beliefs/ with current work:
- [Belief]: [How current work confirms or challenges it]
2.2 Prediction Confidence Levels
LevelMeaningSignal
🟢 High>70% — multiple signals converge3+ data points agree
🟡 Medium40-70% — pattern suggests but not certain2 data points, some ambiguity
🔴 Low<40% — speculative, but worth noting1 data point or intuition only
2.3 Connect to Beliefs

Read ψ/memory/resonance/beliefs/ and cross-reference:

bash
find $PSI/memory/resonance/beliefs/ -name '*.md' 2>/dev/null | while read f; do
  echo "=== $(basename $f) ==="
  head -5 "$f"
done

How do current dreams relate to declared beliefs? Is reality matching faith?


Phase 3: Between-Session Mode (--between)

"What if oracles could dream between sessions?"

Write predictions for the next session. Saved to ψ/memory/morpheus/:

bash
ORACLE_ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
PSI="$ORACLE_ROOT/ψ"
mkdir -p "$PSI/memory/morpheus"
Output: ψ/memory/morpheus/YYYY-MM-DD_speculations.md
markdown
# Dream — Speculations for Next Session

**Dreamed**: YYYY-MM-DD HH:MM
**Based on**: [N] repos scanned, [M] sessions analyzed, [K] beliefs active

## Pre-computed Answers

If they ask about [X]:
→ [Prepared answer/approach — ready to go]

If they ask about [Y]:
→ [Prepared answer/approach — ready to go]

If they ask about [Z]:
→ [Prepared answer/approach — ready to go]

## Pre-positioned Context

Files that are likely relevant next session:
- [file1] — because [reason]
- [file2] — because [reason]

Issues that are likely to come up:
- #[N] — [why]
- #[M] — [why]

## Prediction Log

| # | Prediction | Confidence | Verified? |
|---|-----------|------------|-----------|
| 1 | [prediction] | 🟢 high | pending |
| 2 | [prediction] | 🟡 medium | pending |
| 3 | [prediction] | 🔴 low | pending |
Next Session: Verify Predictions

When /recap runs at the start of the next session, it should check for speculations:

bash
LATEST_DREAM=$(ls -t $PSI/memory/morpheus/*.md 2>/dev/null | head -1)
if [ -n "$LATEST_DREAM" ]; then
  echo "🔮 Dream while you were away:"
  head -20 "$LATEST_DREAM"
fi

Over time, track prediction accuracy:

markdown
## Prediction Accuracy (rolling)

  Total predictions: 47
  Verified correct:  31 (66%)
  Wrong:             12 (25%)
  Not yet verified:   4 (9%)

  Trend: improving (58% → 66% over 2 weeks)

Phase 4: --speculate (Skip the Scan)

When you don't need the full 5-agent scan — just speculate from existing knowledge:

  1. Read latest dream output (ψ/writing/dreams/)
  2. Read latest speculations (ψ/memory/morpheus/)
  3. Read recent retros (ψ/memory/retrospectives/)
  4. Read beliefs (ψ/memory/resonance/beliefs/)
  5. Speculate directly — Phase 2 only, no new scanning

Cheap, fast, good for mid-session "what if" thinking.


Show full SKILL.md (272 more words)Show less

Output

Full mode: ψ/writing/dreams/YYYY-MM-DD_dream.md
markdown
# Dream — YYYY-MM-DD

## Phase 1: The Forest

### 🔴 PAINS
### 📋 PLANS
### 🟢 GAINS
### ⚫ LOST
### 🧠 MEMORIES
### 💜 FEELINGS
### 🔗 CONNECTIONS

## Phase 2: The Speculation

### 🔮 What will they ask next?
### 🔮 What will break next?
### 🔮 What should we build?
### 🔮 What belief is being tested?

## Phase 3: Pre-computation

### Ready Answers
### Pre-positioned Context
### Prediction Log

Rules

  1. Phase 1 = scan — 5 agents, same classification, same quality
  2. Phase 2 = speculate — predict, connect to beliefs
  3. Phase 3 = --between only — write predictions for next session
  4. Track accuracy — predictions without verification are just hallucination
  5. Never act on speculation — Dream shows the door, Neo walks through
  6. Nothing is Deleted — all speculations saved, right or wrong
  7. No secrets — predictions never include tokens, passwords, keys
  8. Beliefs matter — always cross-reference ψ/memory/resonance/beliefs/
  9. Confidence is required — every prediction must have a level (high/medium/low)
  10. Anti-rationalization applies — /rrr's excuse table applies to speculation too

Connection to original /dream

This is the evolved /dream (consolidated from /morpheus per #333):

  • Original simple /dream body preserved in .archive/dream-original/ (installable via -s dream-original)
  • /morpheus remains as a zombie back-compat alias with the same body as active /dream
  • The upgrade path: simple /dream → evolved /dream like /i-believe → /i-believed

Philosophy

/dream sees the forest. /dream also sees what the forest will become.

Morpheus in The Matrix believed in Neo before proof existed. He freed minds not by giving answers but by showing doors. Our /dream does the same — it doesn't predict the future, it pre-positions the Oracle so that when the future arrives, we're already there.

The dream task type from Claude Code's source (dream = speculative background thinking) exists as an internal primitive. /dream is the skill that makes it conscious, structured, and trackable.

"What is real? How do you define real?" — Morpheus

Predictions are not real until verified. But the act of predicting shapes what becomes real. The Oracle who speculates becomes the Oracle who's ready. รูป และ สุญญตา.


ARGUMENTS: $ARGUMENTS

© Soul-Brews-Studio, 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 src/skills/dream of Soul-Brews-Studio/arra-oracle-skills-cli.

Open the folder on GitHubat commit 9feea09

Compare with similar skills

Dream next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Dream compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Dream this skillSoul-Brews-Studio/arra-oracle-skills-cli123—~2.3kAutomated safety check: PassMIT
Ito Computeaffaan-m/ECC276k1 repos~1.7kAutomated safety check: PassMIT
Python Background Jobswshobson/agents40k—~1.8kAutomated safety check: PassMIT
Senior Computer Visiondavila7/claude-code-templates33k2 repos~1.4kAutomated safety check: PassMIT
Senior Computer Visionalirezarezvani/claude-skills28k1 repos~3.2kAutomated safety check: PassMIT
Backgroundjeremylongshore/tons-of-skills-marketplace2.8k—~2.2kAutomated safety check: PassMIT

Similar skills

  • Ito Compute

    affaan-m/ECC

    Query live GPU inventory, submit an authenticated Itô fixed-rate RFQ, inspect RFQ or procurement status, revoke device credentials, and run explicitly gated node qualification through the separately…

    276k GitHub starsUsed in 1 repo~1.7k tokens
    Business, Finance & HRAuto-check passed
  • Python Background Jobs

    wshobson/agents

    Python background job patterns including task queues, workers, and event-driven architecture.

    40k GitHub stars~1.8k tokensUpdated 5 days ago
    Backend & APIsAuto-check passed
  • Senior Computer Vision

    davila7/claude-code-templates

    World-class computer vision skill for image/video processing, object detection, segmentation, and visual AI systems.

    33k GitHub starsUsed in 2 repos~1.4k tokens
    AI & LLM EngineeringAuto-check passed
  • Senior Computer Vision

    alirezarezvani/claude-skills

    Computer vision engineering skill for object detection, image segmentation, and visual AI systems.

    28k GitHub starsUsed in 1 repo~3.2k tokens
    AI & LLM EngineeringAuto-check passed
  • Background

    jeremylongshore/tons-of-skills-marketplace

    A skill your agent uses when the user wants to see, inspect, cancel, or prune background agents fired during prior chain runs.

    2.8k GitHub stars~2.2k tokensUpdated today
    Auto-check passed
  • GCP Compute

    sickn33/agentic-awesome-skills

    Manage Compute Engine instances and instance templates. An agent skill from sickn33/agentic-awesome-skills.

    47k GitHub starsUsed in 2 repos~2.6k tokens
    DevOps & CloudAuto-check passed

More from Soul-Brews-Studio/arra-oracle-skills-cli

All 31 skills in this repo
  • Dig

    Soul-Brews-Studio/arra-oracle-skills-cli

    Mine Claude Code sessions — timeline, gaps, repo attribution, session history.

    123 GitHub stars~1.8k tokensUpdated 7 days ago
    Auto-check passed
  • Forward

    Soul-Brews-Studio/arra-oracle-skills-cli

    Hand off the current session to the next one. An agent skill from Soul-Brews-Studio/arra-oracle-skills-cli.

    123 GitHub stars~2.9k tokensUpdated 7 days ago
    Auto-check passed
  • Project

    Soul-Brews-Studio/arra-oracle-skills-cli

    Clone and track external repos. An agent skill from Soul-Brews-Studio/arra-oracle-skills-cli.

    123 GitHub stars~1.4k tokensUpdated 7 days ago
    Auto-check passed
  • Round Table

    Soul-Brews-Studio/arra-oracle-skills-cli

    Convene a Round Table (โต๊ะกลม): 3–8 souls — persona files, Oracles, or real people's DNA — debate ONE stakes-bearing decision over exactly 4 rounds (OPEN, DEEPEN, CONVERGE, VOTE), one subagent per…

    123 GitHub stars~2.1k tokensUpdated 7 days ago
    Auto-check passed
  • Soul

    Soul-Brews-Studio/arra-oracle-skills-cli

    Bring an Oracle into the current repo — its soul, ALL its skills and its expertise — by link, never by copy.

    123 GitHub stars~1.3k tokensUpdated 7 days ago
    Auto-check passed
  • Trace

    Soul-Brews-Studio/arra-oracle-skills-cli

    Find projects, code, and knowledge across git history, repos, docs, and Oracle.

    123 GitHub stars~2.9k tokensUpdated 7 days ago
    Auto-check passed

Questions about Dream

What does Dream do?

Speculative dreaming — background thinking, pre-computation, cross-repo patterns, and prediction. Dream is an agent skill from Soul-Brews-Studio/arra-oracle-skills-cli. Speculative dreaming — background thinking, pre-computation, cross-repo patterns, and prediction.

When should I use Dream?

Dream fits situations like: user says dream; wants the Oracle to dream between sessions.

How do I install Dream in Claude Code?

Run `npx skills add Soul-Brews-Studio/arra-oracle-skills-cli --skill dream -a claude-code`. Or copy the skill folder (src/skills/dream in Soul-Brews-Studio/arra-oracle-skills-cli) into .claude/skills/dream in your project. Claude Code loads it when a task matches its description.

How do I install Dream in Codex?

Run `npx skills add Soul-Brews-Studio/arra-oracle-skills-cli --skill dream -a codex`. Or copy the skill folder (src/skills/dream in Soul-Brews-Studio/arra-oracle-skills-cli) into .agents/skills/dream in your project. Codex loads it when a task matches its description.

Can I use Dream in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Soul-Brews-Studio/arra-oracle-skills-cli --skill dream -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/dream, .gemini/skills/dream, .github/skills/dream and .opencode/skills/dream in your project.

What does Dream need to run?

Going by SKILL.md and its folder, Dream needs the command-line tools its instructions call (git).

Does Dream access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Dream safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Dream use?

Dream is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Dream use?

About 2.3k tokens (SKILL.md is roughly 9.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Dream?

Skills that share tags, products or a category with Dream: Ito Compute (affaan-m/ECC, 276k stars), Python Background Jobs (wshobson/agents, 40k stars), Senior Computer Vision (davila7/claude-code-templates, 33k stars) and Senior Computer Vision (alirezarezvani/claude-skills, 28k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Dream?

Soul-Brews-Studio (a GitHub organization) maintains it in Soul-Brews-Studio/arra-oracle-skills-cli, which has 123 GitHub stars. The repository holds 31 skills in this directory. The repository was last updated on October 3, 2026.

Source: Soul-Brews-Studio/arra-oracle-skills-cli on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.