openclaw-agentmemory-plugin Package Skill

Package: @epicdm/openclaw-agentmemory-plugin Repository: epic-flowstate-community Path: packages/openclaw-agentmemory-plugin Version: 0.2.2

Use this published Dojo skill as the package-level entrypoint. For detailed workflows, use the package-local skills listed below.

Primary Package Workflow

FlowState OpenClaw AgentMemory Plugin

Status: Active Purpose: Guide plugin entry and OpenClaw lifecycle work. Scope: packages/openclaw-agentmemory-plugin. Trigger: Need to modify plugin registration, config, manifest, or hook wiring. Input: OpenClaw API, plugin config, env vars, and AgentMemory endpoint. Output: Plugin entry that correctly wires AgentMemory runtime into OpenClaw.

Workflow

1. Keep openclaw.plugin.json aligned with package.json.
2. Keep index.ts as the plugin entry listed in openclaw.extensions.
3. Map both snake_case and camelCase plugin config fields.
4. Register memory capability only when api.registerMemoryCapability exists.
5. Wire before_agent_start and agent_end hooks to the runtime.

Verification

yarn workspace @epicdm/openclaw-agentmemory-plugin test --runInBand --no-cache
yarn workspace @epicdm/openclaw-agentmemory-plugin typecheck
yarn workspace @epicdm/openclaw-agentmemory-plugin build

Created: 2026-06-29

Package Skill Inventory

@epicdm/openclaw-agentmemory-plugin Skill Inventory

Review Gate

@epicdm/openclaw-agentmemory-plugin Skills Review

Verdict: Pass with follow-ups.

Skills cover entry registration, runtime config, prompt memory recall, observation writes, fallback behavior, and verification commands.

Verification

• ASCII and skill frontmatter checks passed.
• yarn workspace @epicdm/openclaw-agentmemory-plugin test --runInBand --no-cache passed: 1 suite, 4 tests.
• yarn workspace @epicdm/openclaw-agentmemory-plugin typecheck passed.
• yarn workspace @epicdm/openclaw-agentmemory-plugin build passed.
• yarn workspace @epicdm/openclaw-agentmemory-plugin lint passed.

Follow-Ups

• Rebuild generated .flowstate/dojo content from the reviewed local skills before publishing.

Start with this skill to understand the package purpose, then use the package-local focused skills for specific workflows. Keep this Dojo version aligned with packages/openclaw-agentmemory-plugin/package.json. Do not treat generated feature lists as lessons; use workflow evidence from CODE-REVIEW.md, SKILL-INVENTORY.md, and REVIEW.md.