[Feature]: Ship first-party Skills with Memoria for explicit MCP tool invocation

[loveRhythm1990]

Is there an existing issue for the same feature?

• I have checked the existing issues.

Is your feature request related to a problem?

NA

Describe the feature you'd like

Requirements

Memoria MCP tools (memory_governance, memory_consolidate, memory_search, etc.) work well when the agent calls them automatically, but there is no reliable way for users to trigger a specific tool on demand.

We validated a pattern: a Skill that maps a slash command (e.g. /governance) to a Memoria MCP tool. The Skill instructs the agent to call the tool immediately and report results in a consistent format. It works, but users must write and maintain these Skills themselves.

We want Memoria to ship first-party Skills out of the box, so common memory operations are one command away without manual setup.

Initial set:

Skill	MCP tool
/governance	memory_governance
/consolidate	memory_consolidate
/mem-search	memory_search
/mem-store	memory_store
/reflect	memory_reflect

Each Skill should: call the tool first (no skipping), document params and cooldowns, and use a standard output template.

---

Possible Implementation

Option A (recommended): bundle Skills on install

When users set up Memoria (e.g. memoria mcp --tool cursor or memoria init), optionally install bundled Skills:

memoria init --with-skills

Copy templates from memoria/templates/skills/ into the project (e.g. .cursor/skills/). Version and update with Memoria releases.

Option B: MCP Prompts as complement

Expose the same workflows as MCP Prompts for clients that support /prompt_name, sharing the same tool mapping and output schema as the Skills.

Option C: hybrid

Skills for rich UX + MCP Prompts where supported; single source of truth for tool mapping in the Memoria repo.

---

Acceptance Criteria

• Ship ≥5 first-party Skills matching the table above
• One-command install (e.g. memoria init --with-skills)
• Each Skill documents MCP params, cooldowns, and expected response
• README with a /governance example

Implementation / design notes (optional)

Additional information

No response