dreaming
Creative exploration during quiet hours. Turns idle heartbeat time into freeform thinking — hypotheticals, future scenarios, reflections, unexpected connections. Use when you want your agent to do something meaningful during low-activity periods instead of just returning HEARTBEAT_OK. Outputs written to files for human review later (like remembering dreams in the morning).
What this skill does
# Dreaming
Creative, exploratory thinking during quiet hours. Not task-oriented work — freeform associative exploration that gets captured for later review.
## Setup
### 1. Configure quiet hours and topics
Edit `scripts/should-dream.sh` to customize:
- **QUIET_START / QUIET_END** — when dreaming can happen (default: 11 PM - 7 AM)
- **TOPICS array** — categories of exploration (see defaults for examples)
### 2. Create state and output directories
```bash
mkdir -p data memory/dreams
```
### 3. Add to HEARTBEAT.md
Add this section to your heartbeat routine (during quiet hours):
```markdown
## Dream Mode (Quiet Hours Only)
Check if it's time to dream:
\`\`\`bash
DREAM_TOPIC=$(./scripts/should-dream.sh 2>/dev/null) && echo "DREAM:$DREAM_TOPIC" || echo "NO_DREAM"
\`\`\`
**If DREAM_TOPIC is set:**
1. Parse the topic (format: `category:prompt`)
2. Write a thoughtful exploration to `memory/dreams/YYYY-MM-DD.md`
3. Keep it genuine — not filler. If the well is dry, skip it.
4. Append to the file if multiple dreams that night
```
## How It Works
The `should-dream.sh` script acts as a gate:
1. Checks if current time is within quiet hours
2. Checks if we've already hit the nightly dream limit
3. Rolls dice based on configured probability
4. If all pass: returns a random topic and updates state
5. If any fail: exits non-zero (no dream this heartbeat)
State tracked in `data/dream-state.json`:
```json
{
"lastDreamDate": "2026-02-03",
"dreamsTonight": 1,
"maxDreamsPerNight": 1,
"dreamChance": 1.0
}
```
## Writing Dreams
When the script returns a topic, write to `memory/dreams/YYYY-MM-DD.md`:
```markdown
# Dreams — 2026-02-04
## 01:23 — The Future of X (category-name)
[Your exploration here. Be genuine. Think freely. Make connections.
This isn't a report — it's thinking out loud, captured.]
```
**Guidelines:**
- One dream = one topic, explored thoughtfully
- Timestamp each entry
- Append if multiple dreams in one night
- Skip if you have nothing worth saying — forced dreams are worthless
- This is for your human to review later, like reading a journal
## Customizing Topics
**Option A: Config file (recommended)** — Create `data/dream-config.json`:
```json
{
"topics": [
"future:What could this project become?",
"creative:A wild idea worth exploring",
"reflection:Looking back at recent work"
]
}
```
This keeps your customizations outside the skill directory (safe for skill updates).
**Option B: Edit script directly** — Modify the `DEFAULT_TOPICS` array in `should-dream.sh`. Format: `category:prompt`
Default categories:
- `future` — What could [thing] become?
- `tangent` — Interesting technology or concepts worth exploring
- `strategy` — Long-term thinking
- `creative` — Wild ideas that might be crazy or brilliant
- `reflection` — Looking back at recent work
- `hypothetical` — What-if scenarios
- `connection` — Unexpected links between domains
Add domain-specific topics relevant to your work. The prompt should spark genuine exploration, not busywork.
## Tuning
In `data/dream-state.json`:
Add domain-specific topics relevant to your work. The prompt should spark genuine exploration, not busywork.
## Tuning
In `data/dream-state.json`:
- **maxDreamsPerNight** — cap on dreams per night (default: 1)
- **dreamChance** — probability per check (default: 1.0 = guaranteed if under limit)
Lower `dreamChance` for more sporadic dreaming. Raise `maxDreamsPerNight` for more prolific nights.
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.