LLMSkillHub
mirroredCopied from joshuaswarren/remnic under MIT. joshuaswarren has not claimed this page, and is not affiliated with LLM SkillHub.

joshuaswarren / remnic-memory-workflow

Shared memory workflow for Claude Code agents connected to Remnic — recall before acting, observe during work, remember at the end. Trigger phrases include "what do you remember about", "save this for later", "any context from last time".

0
SKILL.md
3.2 KB

When to use

Use this skill as the default playbook whenever Claude Code picks up a task that could benefit from prior context, or when the user explicitly asks the agent to remember or recall something. The individual remnic-recall, remnic-remember, remnic-search, remnic-entities, and remnic-status skills implement the detailed steps.

Triggers:

  • New ticket, branch, or task begins.
  • "What do you remember about …"
  • "Save this for later."
  • Long-running turn produces a durable outcome worth capturing.

Inputs

  • Current user request (natural language).
  • Optional: active project path, ticket number, branch name.
  • Optional: topic keywords surfaced earlier in the session.

Procedure

  1. Recall first. Call remnic_recall with a concise natural-language query built from the user's request. Pull 3–8 results and filter for relevance.
  2. Mention relevant memories briefly to the user when they change the plan; otherwise use them as quiet context.
  3. Observe during work. For significant tool results (Write/Edit/MultiEdit, Bash exits, test output) call remnic_observe so Remnic keeps its ambient context fresh.
  4. Deep search on demand. If remnic_recall missed something the user insists exists, fall back to remnic_lcm_search with a more literal phrase.
  5. Browse entities. When the user names a project, person, or concept, call remnic_entity_get to pull facts and relations.
  6. Remember at the end. Before ending the turn, store durable decisions, preferences, and findings via remnic_memory_store.

Efficiency plan

  • One broad recall beats several narrow ones.
  • Skip recall for trivially local tasks (formatting, arithmetic, mechanical refactors).
  • Reuse recall results within the same turn.
  • Store each memory once; update rather than duplicate.

Pitfalls and fixes

  • Pitfall: Storing transient state. Fix: Only store facts with durable value.
  • Pitfall: Leaking secrets. Fix: Redact credentials and tokens before calling remnic_memory_store.
  • Pitfall: Flooding the user with recalled context. Fix: Summarize in 1–3 bullet points.
  • Pitfall: Forgetting the final write. Fix: Make "remember the decision" the last step of any non-trivial turn.

Verification checklist

  • [ ] Recall was attempted before the agent committed to a plan.
  • [ ] User-facing mentions of recalled context were concise and relevant.
  • [ ] Durable decisions and findings were stored via remnic_memory_store.
  • [ ] No secrets, credentials, or transient state were written.
  • [ ] Canonical remnic_* tool names were used over legacy aliases.

Tool names: this skill uses the canonical remnic_* MCP tools. Legacy engram_* aliases remain accepted during v1.x for backward compatibility.