templates/skills/modules/synap/SKILL.md
Use MCP Synap for persistent task and data storage across context windows.
npx skillsauth add hivellm/rulebook SynapInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
CRITICAL: Use MCP Synap for persistent task and data storage across context windows.
synap_kv_set(key, value, ttl?) # Store with optional TTL
synap_kv_get(key) # Retrieve
synap_kv_delete(key) # Remove
synap_kv_scan(prefix) # List by prefix
synap_queue_publish(queue, message, priority) # Add to queue
synap_queue_consume(queue) # Process from queue
synap_pubsub_publish(topic, message) # Broadcast
synap_stream_publish(room, event) # Stream events
Pattern: "task:<feature>:<subtask-id>"
synap_kv_set("task:auth:login", JSON.stringify({
status: "in_progress",
tests: ["test_login_success"],
coverage: 95.2
}))
Pattern: "session:<id>:<data-type>"
synap_kv_set("session:abc:current-file", "/src/auth.ts")
synap_kv_set("session:abc:todo-list", JSON.stringify([...]))
Pattern: "test:<suite>:<timestamp>"
synap_kv_set("test:integration:latest", JSON.stringify({
passed: 42,
failed: 0,
coverage: 96.5
}), 86400) // TTL: 24 hours
✅ DO:
❌ DON'T:
research
Create structured analyses with numbered findings, execution plans, and task materialization
research
Author a rulebook task spec interactively — research, draft, ask the user clarifying questions, confirm, then create the tasks in rulebook ready for /rulebook-driver. Use when the user wants to plan/spec a feature before implementing.
development
Behavioral guidelines to reduce common LLM coding mistakes — overcomplication, sloppy refactors, hidden assumptions, weak goals. Use when writing, reviewing, or refactoring code. Auto-applies; invoke explicitly via /karpathy-guidelines or 'follow karpathy discipline'.
data-ai
Autonomous AI agent loop for iterative task implementation (@hivehub/rulebook ralph)