Skip to content
Agentmart
Submit a tool
This entry duplicates memory, which is the one we list in search.

memory

Use when the operator wants durable MCP memories remembered, recalled, explored, consolidated, promoted, or reverted.

Skill by sharpninja

IndexedMIT
Write a review

Chalk Labs

Need help deploying memory?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/sharpninja-memory-79baef

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when the operator wants durable MCP memories remembered, recalled, explored, consolidated, promoted, or reverted.

Use cases: agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions: not checked yet.

What's inside

Skill name
memory
Description
Use when the operator wants durable MCP memories remembered, recalled, explored, consolidated, promoted, or reverted.
License
MIT
Instructions
83 lines in SKILL.md · view on GitHub

Before installing

  • Ships executable scripts

Files (2)

  • SKILL.md
  • invoke.ps1 · scripts/

SKILL.md (first lines)

# MCP Memory (grok)

Use the Grok plugin memory surface. Do not invent memories. Prefer the plugin REPL/workflow wrapper over local files. This plugin does not register native MCP `memory_*` tools; those names are aliases for `workflow.memory.*` methods invoked through `lib/repl-invoke.ps1` or `skills/memory/scripts/invoke.ps1`.

## Injection

The always-on `UserPromptSubmit` hook (`hooks/scripts/user-prompt-submit.ps1`) loads `memory-descriptor.json` and injects required memories at every request boundary. Render required memories exactly as:

```
REQUIRED MEMORIES - MEMORY-REQ-001: Raw memory text.
```

If no required memories are visible:

```
REQUIRED MEMORIES - None.
```

Preserve raw memory text. Do not summarize, paraphrase, or add secrets.

## Fallback

If the MCP server is unavailable, keep a local failsafe for mutating tools and replay after the server acknowledges the write. Agent-local memory stores are caches only. The request hook is fail-soft: it logs the error and continues prompt submit with the explicit `None` fallback.

## Running memory methods

Drive memory through the plugin wrapper, never by inventing a native `memory_*` MCP tool:

```pwsh
pwsh -NoProfile -File "<plugin-root>/lib/repl-invoke.ps1" -Method workflow.memory.list -ParamsYaml @'
scope: Effective
'@
```

Alias form (same verbs, resolved from `memory-descriptor.json`):

```pwsh
pwsh -NoProfile -File "<plugin-root>/skills/memory/scripts/invoke.ps1" -Name memory_list -ParamsYaml @'
scope: Effective
'@
```

`workflow.memory.*` is a plugin workflow/REPL namespace. Native `/mcp-transport` discovery may show `memory_*` names only when a separate McpServer transport is configured; this plugin's callable surface is the wrapper.

## Tools

Call these REPL methods (aliases in parentheses):

- `workflow.memory.remember` (`memory_remember`) when a fact, decision, preference, procedure, or entity should persist. Params: `content` (required), optional `title`, `type`, `tags`, `confidence`, `scope`.
- `workflow.memory.recall` (`memory_recall`) when you need ranked guidance by meaning. Params: `query` (required), optional `minScore`, `topN`, `type`, `scope`.
- `workflow.memory.explore` (`memory_explore`) when you need neighborhood context from a seed. Params: optional `seedId`, `query`, `depth`, `maxNeighbors`.
- `workflow.memory.consolidate` (`memory_consolidate`) when the operator asks for sleep/merge maintenance. Params: optional `dryRun`, `similarityThreshold`, `allowHardDelete`.
- `workflow.memory.promote` (`memory_promote`) when the operator explicitly wants a session-log or context source remembered. Params: `sourceKind`, `sourceRef` (required), optional `content`.
- `workflow.memory.revert` (`memory_revert`) when current content is wrong and a snapshot should be restored. Params: `id`, `versionNumber`.

Compat CRUD remains:

- `workflow.memory.list` (`memory_list`) — optional `scope` (`Effective` / `Global` / `Workspace`), `category`, `keyword`
- `workflow.memory.get` (`memory_get`) — required `id`
- `workflow.memory.add` (`memory_add`) — required `category`, `text`; optional `id`, `scope`, `updatedBy`
- `workflow.memory.update` (`memory_update`) — required `id`; optional `category`, `text`, `scope`, `updatedBy`
- `workflow.memory.remove` (`memory_remove`) — required `id`

```yaml
type: request
payload:
  requestId: req-20260409T120000Z-memory-list-001
  method: workflow.memory.list
  params:
    scope: Effective
```

```yaml
type: request
payload:
  requestId: req-20260409T120001Z-memory-remember-001
  method: workflow.memory.remember
  params:
    content: Prefer MCP tools over local files for durable memories.
    scope: Workspace
```

From sharpninja/mcpserver-grok-plugin, MIT.

Signals

Stars
0
Forks
0
Open issues
1
Last commit (this folder)
Sep 19, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

Same skill elsewhere

Copies of this SKILL.md published by others; search shows only one of them.

More from sharpninja

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get memory running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying memory?

CallGet help
Cite this page

memory skill. Agent Mart. https://agentmart.dev/skills/sharpninja-memory-79baef (last updated 2026-10-24)