What makes mem9.ai notable?

2.2K downloads. Trust evidence available

How should mem9.ai be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

mem9.ai

Persistent cloud memory plugin for OpenClaw. This document routes setup, troubleshooting, and uninstall flows and defines config boundaries.

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

2.2k

Updated

Oct 9, 2026

Version

1.0.48

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 2.2K downloads reported by the source. Last updated 10/9/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 9, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 9, 2026
Adoption signal: 2.2K downloadsadoption · observed Oct 9, 2026
Latest release: 1.0.48release · observed May 6, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s170h6wt11ced8trc7m9dktafs8416mg:mem9-ai

Install using `clawhub skill install s170h6wt11ced8trc7m9dktafs8416mg:mem9-ai` in an isolated environment before connecting it to live workloads.
No published capability contract is available yet, so validate auth and request/response behavior manually.
Review the upstream CLAWHUB listing at https://clawhub.ai/c4pt0r/mem9-ai before using production credentials.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-c4pt0r-mem9-ai/snapshot"

Documentation

CLAWHUB

143,933 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: mem9
version: 1.0.49
homepage: "https://mem9.ai/openclaw-memory"
description: "Persistent cloud memory plugin for OpenClaw. This document routes setup, troubleshooting, and uninstall flows and defines config boundaries."
keywords: ["mem9", "openclaw memory", "persistent memory", "cloud memory", "memory plugin", "setup mem9", "install mem9", "reconnect mem9", "uninstall mem9"]
user-invocable: true
metadata: {"openclaw":{"emoji":"🧠","homepage":"https://mem9.ai/openclaw-memory","primaryEnv":"MEM9_API_KEY"}}
---

# mem9

mem9 is the persistent cloud memory plugin for OpenClaw. It keeps durable memory across sessions and machines.

This top-level `SKILL.md` is a routing and safety document. It is the entry point for setup, troubleshooting, and uninstall flows. It is not the canonical reference for runtime request behavior.

## What This Skill Does

- Route first-time install, reconnect, config review, and mainland China / Alibaba Cloud npm install handling to `SETUP.md`.
- Route setup failures, recovery work, and broken local state to `TROUBLESHOOTING.md`.
- Route local mem9 removal on this machine to `UNINSTALL.md`.
- Preserve filtered `utm_*` params from a remote `SKILL.md` URL through the create-new setup flow and first restart.
- Install mem9 only through `openclaw plugins install @mem9/mem9`; never substitute `npm install -g @mem9/mem9`, because OpenClaw does not discover global npm packages as plugins.
- Prevent invalid compatibility fallbacks: never write `plugins.entries.mem9.apiKey`; only `plugins.entries.mem9.config.apiKey` is valid.
- Configure OpenClaw 4.23+ hook permission with `plugins.entries.mem9.hooks.allowConversationAccess = true` so mem9 can read conversation messages during `agent_end`.
- Keep create-new verification deterministic: if the first post-restart return message does not produce a key, run the one local OpenClaw agent trigger from `SETUP.md` in that same resumed turn; never ask for repeated `hi` messages.
- In create-new mode, do not treat missing `plugins.entries.mem9.config.apiKey` as failure. The generated key is verified from mem9 logs or local provision state until the user explicitly reconnects with it.
- Enforce the setup success onboarding handoff: every successful create-new or reconnect setup must end with the template that starts `✅ Your mem9 API key is connected.` followed by `🧭 WHAT YOU CAN DO NEXT`; never replace it with a brief success summary.
- Keep API keys out of local memory/history notes during setup. Do not write the full key or a key preview into `MEMORY.md`, `memory/*.md`, or equivalent local logs; show the key only in the final onboarding handoff.
- Keep restart notices honest: only claim post-restart verification has resumed after that work has actually started.
- Keep history import manual. Do not scan or upload local memory or session files from this skill.

## Source / Runtime Authority

- Official source: [mem9-ai/mem9](https://github.com/mem9-ai/mem9) and [mem9.ai](https://mem9.

_meta.json

{
  "ownerId": "kn7cn03y7y2x2cqfd2wd4ybtzn82hgss",
  "slug": "mem9-ai",
  "version": "1.0.48",
  "publishedAt": 1778085185933
}

SETUP.md

# mem9 Setup

Use this file only for first-run mem9 onboarding or reconnect.

## First Reply

When this file is read, send only the choice prompt below. Do not include version checks, config paths, config keys, dry-run details, restart counts, or approval phrases in the first reply.

> I can help you set up mem9.
>
> Choose one path:
> 1. Create a new mem9 API key during setup
> 2. Reconnect an existing mem9 API key
>
> If you choose reconnect, paste your existing API key.

## Hard Rules

- Only edit the exact `openclaw.json` keys listed in `SKILL.md`.
- Do not upload local history, memory files, or session files during setup.
- Do not use raw mem9 API calls, direct web-request provisioning, or scripted shell patch commands as a substitute for the plugin flow.
- Do not write the generated or provided API key, even a preview, into local memory files, history notes, `MEMORY.md`, `memory/*.md`, session summaries, or setup logs. Show the key only in the final onboarding handoff.
- Install only with `openclaw plugins install @mem9/mem9`, optionally prefixed by one-shot `NPM_CONFIG_REGISTRY=https://registry.npmmirror.com`. Never substitute `npm install -g @mem9/mem9`, `npm install @mem9/mem9`, `npx`, or a copied npm package.
- Do not point `plugins.slots.memory` at `mem9` until the OpenClaw plugin install succeeds and OpenClaw recognizes plugin id `mem9`.
- Reconnect and create-new are separate branches. After the user chooses one, do not switch unless the user explicitly changes their choice.
- The only valid secret config path is `plugins.entries.mem9.config.apiKey`. Never write `plugins.entries.mem9.apiKey`.
- On OpenClaw `4.23+` / `2026.4.22+`, write `plugins.entries.mem9.hooks.allowConversationAccess = true` as a sibling of `enabled` and `config`. On older versions that reject it, omit it and tell the user full automatic conversation upload requires an OpenClaw upgrade.
- In create-new mode, generate `PROVISION_TOKEN` once after the branch is selected, write it before the first restart, and keep it stable until provisioning succeeds on this machine.
- If the remote `SKILL.md` URL has query params whose keys start with `utm_`, preserve those exact filtered pairs as `REMOTE_SKILL_UTM_PARAMS` and write them only in create-new mode under `plugins.entries.mem9.config.provisionQueryParams`.
- Show one dry-run preview and get one approval for the disclosed install, config, restart, and verification flow. Ask for another approval only if scope materially changes.
- Do not ask the user for repeated `hi` messages while the only missing step is create-new provisioning. Use the one local trigger in this document.
- If install fails, stale local plugin state blocks install, config is already invalid, reconnect auto-provisions a new key, multiple create-new keys appear, or the documented checks cannot be satisfied, stop and use `TROUBLESHOOTING.md`.

## Required Success Handoff

This template is intentionally near the top so remote reads see it even if later se

skill-card.md

## Description:

Persistent cloud memory plugin for OpenClaw. This document routes setup, troubleshooting, and uninstall flows and defines config boundaries.

This skill is ready for commercial/non-commercial use.

## Publisher:

[c4pt0r](https://clawhub.ai/user/c4pt0r)

### License/Terms of Use:

MIT-0

## Use Case:

Developers and OpenClaw users use this skill to install, reconnect, troubleshoot, or remove mem9 as persistent cloud memory across sessions and machines. It guides configuration changes, plugin installation, verification, and cleanup while preserving API-key and local-history boundaries.

### Deployment Geography for Use:

Global

## Known Risks and Mitigations:

Risk: mem9 acts as persistent cloud memory and may receive conversation-derived content when conversation access is enabled.

Mitigation: Get explicit user consent before enabling conversation access, disclose the configured API URL, and do not upload local history or memory files during setup.

Risk: The setup stores and reuses a mem9 API key, which could expose account memory if copied into logs, history, or chat transcripts.

Mitigation: Keep the API key only in the documented plugin config path and user handoff, avoid local memory or history notes, and store the key in a secure vault.

Risk: The documented install command installs an unpinned plugin package.

Mitigation: Use a pinned, reviewed plugin version with integrity verification before deployment where release control matters.

Risk: Incorrect OpenClaw config paths or setup ordering can leave the memory plugin unavailable or the gateway unhealthy.

Mitigation: Restrict edits to the documented OpenClaw keys, install the plugin before switching the memory slot, read back config before restart, and use the troubleshooting flow when verification fails.

## Reference(s):

- [mem9 OpenClaw Memory](https://mem9.ai/openclaw-memory)
- [mem9 Website](https://mem9.ai/)
- [mem9 GitHub Project](https://github.com/mem9-ai/mem9)
- [OpenClaw Plugin README](https://github.com/mem9-ai/mem9/blob/main/openclaw-plugin/README.md)
- [mem9 Memory Dashboard](https://mem9.ai/your-memory/)

## Skill Output:

**Output Type(s):** [Guidance, Markdown, Shell commands, Configuration]

**Output Format:** [Markdown instructions with inline shell commands and configuration paths]

**Output Parameters:** [1D]

**Other Properties Related to Output:** [Routes setup, troubleshooting, uninstall, and post-setup memory-use workflows for OpenClaw.]

## Skill Version(s):

1.0.48 (source: server release metadata; artifact frontmatter says 1.0.49)

## Ethical Considerations:

Users should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.

TROUBLESHOOTING.md

# mem9 Troubleshooting

Use this file for reconnect, setup failures, uninstall failures, and dashboard/login confusion after the main mem9 flow.

## Quick Checks

Confirm these first:

- `plugins.slots.memory` is set to `mem9`
- `plugins.entries.mem9.enabled` is `true`
- `plugins.entries.mem9.config.apiUrl` points to the intended mem9 API
- `plugins.entries.mem9.config.apiKey` is present for the steady-state reconnect flow
- In reconnect mode, the read-back value of `plugins.entries.mem9.config.apiKey` exactly matches the user's original key before the first restart
- On OpenClaw `4.23+` / `2026.4.22+`, `plugins.entries.mem9.hooks.allowConversationAccess` is `true`
- On OpenClaw `>= 2.2.0`, `plugins.allow` includes `mem9`

## Common Issues

### Plugin Does Not Load

- Re-check the memory slot and enabled flag
- Re-check that the mem9 package was installed successfully
- Re-check that the config only edits the exact mem9 keys and does not corrupt unrelated JSON

### Install Hit `plugin not found: mem9`

- Treat this as setup ordering failure, not as an API-key problem
- Re-check whether the flow wrote `plugins.slots.memory = "mem9"` before `openclaw plugins install @mem9/mem9` succeeded
- Re-check whether `plugins.entries.mem9` or `plugins.allow += "mem9"` were also written before install success
- If yes, fix the ordering bug in the setup flow: install first, then apply all disclosed mem9 config in one contiguous update
- If the machine is already stuck in that invalid state, temporarily restore only the currently valid memory slot and remove only the premature mem9 wiring, then install the plugin, then re-apply the approved mem9 config
- Do not require a fresh extra approval for that missing-plugin recovery when it is only undoing the same mem9 wiring disclosed in the original setup scope

### Create-New Flow Did Not Return A Key

- Make sure the first restart happened with `plugins.entries.mem9.config.apiKey` absent
- Make sure `plugins.entries.mem9.config.provisionToken` was present and stable before that restart
- Make sure the first resumed user message after that restart actually reached the plugin hook path
- Make sure that resumed turn produced exactly one generated key in logs
- If the first resumed turn did not produce a key, stop the first-run flow and ask the user whether to retry the clean create-new flow or switch to reconnect with an existing API key

### Create-New Flow Did Not Carry Expected UTM Params

- If the setup started from a remote `SKILL.md` URL with `utm_*`, re-check whether `plugins.entries.mem9.config.provisionQueryParams` was disclosed in the dry-run preview
- Re-check the config read-back from before the first restart
- The pre-restart read-back must exactly match the filtered `utm_*` map from the remote `SKILL.md` URL
- If `provisionQueryParams` was absent, partial, renamed, or different before the first restart, do not assume the generated key was attributed correctly
- In that case, stop the happy-path create-new

Github ReposUpdated 5h agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.