---

OpenClaw AI Agent Skill

---

name: council-memory description: "Persistent memory layer for Council Room — reads, writes, and searches agent memory across sessions. Use when saving important decisions/facts to long-term memory, recalling past sessions, updating MEMORY.md or daily notes, searching the Obsidian vault for context, or injecting memory into agent prompts so agents never forget who Lucy is or what DataFlow is building. Replaces the goldfish memory problem."

Council Memory Skill

Gives Council Room agents persistent memory across sessions — like Kora's shared memory, but self-hosted and unlimited.

Memory Architecture

Short-term  → memory/YYYY-MM-DD.md      (daily raw notes)
Long-term   → MEMORY.md                  (curated facts)
Agent facts → vault/00_System_Core/      (Obsidian vault)
Council KB  → council_memory.json        (shared agent facts)

Paths:

• Workspace: /home/openclaw/.openclaw/workspace/
• Vault: /home/openclaw/livesync-bridge/vault/
• Council memory file: /home/openclaw/council-room/data/council_memory.json
• Daily notes: /home/openclaw/.openclaw/workspace/memory/YYYY-MM-DD.md

Core Operations

Read Memory (before answering anything personal)

python3 ~/.openclaw/workspace/skills/council-memory/scripts/memory_op.py read

Write a Fact

python3 ~/.openclaw/workspace/skills/council-memory/scripts/memory_op.py write \
  --fact "Lucy decided to target bookkeepers with 5+ years experience" \
  --category "decisions"

Search Memory

python3 ~/.openclaw/workspace/skills/council-memory/scripts/memory_op.py search \
  --query "bookkeeper validation"

Daily Note Append

python3 ~/.openclaw/workspace/skills/council-memory/scripts/memory_op.py daily \
  --note "Rafael standup: OCR accuracy at 87%, needs 8% improvement"

Inject Context for Agent

python3 ~/.openclaw/workspace/skills/council-memory/scripts/memory_op.py context \
  --agent "rafael"

Returns structured context string to prepend to agent prompts.

When to Use Each Memory Layer

Situation	Where to Write
Important decision made	MEMORY.md + council_memory.json
Daily event/progress	memory/YYYY-MM-DD.md
Agent-specific lesson	vault/Lessons/<agent>.md
Lucy preference discovered	MEMORY.md under "How Lucy Works"
Metric/KPI update	council_memory.json categories.kpis
Technical decision	MEMORY.md under "Technical Decisions"

Auto-Context Injection

When starting a new session or agent chat, call context op to get:

• Who Lucy is (one paragraph)
• Current DataFlow status
• Last 3 decisions made
• Active KPIs

This prevents agents from giving generic answers — they always know:

• Lucy = SaaS founder, Indonesia, UTC+8, DataFlow = AI OCR for bookkeepers
• Current sprint goals
• Budget constraints
• Council structure

council_memory.json Schema

See references/schema.md for full schema. Key categories:

• facts — persistent truths about Lucy/DataFlow
• decisions — logged decisions with date + rationale
• kpis — current metrics (OCR accuracy, bookkeepers found, etc.)
• lessons — what agents learned from past sessions

Installation

cp -r council-memory/ ~/.openclaw/workspace/skills/council-memory/

License

MIT © Sentra Technology