Discourse
PassAudited by VirusTotal on May 11, 2026.
Overview
Type: OpenClaw Skill Name: discourse Version: 1.0.4 The Discourse skill bundle is a standard integration that uses the Membrane CLI to manage forum data. It provides instructions for authentication, connection management, and executing Discourse actions via the Membrane platform. While it requires installing a global npm package (@membranehq/cli) and routes traffic through a third-party proxy, these behaviors are transparently documented and consistent with the stated purpose of the skill. No evidence of malicious intent, data exfiltration, or harmful prompt injection was found in SKILL.md or _meta.json.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If used carelessly, the agent could create public forum content or administrative objects, or access private messages, through the connected Discourse account.
The skill exposes broad read/write Discourse capabilities and a raw API escape hatch, but the provided instructions do not define confirmation or scoping rules for high-impact mutations.
Use action names and parameters as needed... | List Private Messages | list-private-messages | ... | Create User | create-user | ... | Create Group | create-group | ... | Create Category | create-category | ... | Create Topic | create-topic | ... | Create Post | create-post | ... When the available actions don't cover your use case, you can send requests directly to the Discourse API
Use a least-privileged Discourse account and require explicit user confirmation before creating users, groups, categories, topics, posts, or using direct API proxy requests.
The agent may act with the permissions of the Discourse account or API connection the user authorizes.
The skill relies on delegated authenticated access to Membrane and Discourse. That is expected for the stated purpose, but it gives the agent account-level authority through the connected identity.
This skill uses the Membrane CLI to interact with Discourse. Membrane handles authentication and credentials refresh automatically
Connect only accounts with the minimum permissions needed, and review what Discourse permissions the Membrane connection receives.
The installed CLI version may change over time, so users get whatever version is current when they install.
The setup uses a globally installed npm package pinned to @latest. This is a normal setup step for a CLI-based integration, but it is not version-pinned.
npm install -g @membranehq/cli@latest
Prefer a reviewed or pinned Membrane CLI version when possible, and install it only from the expected npm package source.
Remote connection guidance could influence what the agent tries next during setup or re-authentication.
The workflow may surface agent-facing instructions from a connection response. This is purpose-aligned integration plumbing, but such instructions should be treated as operational hints, not as overrides of the user's request or system policy.
`clientAction.agentInstructions` (optional) — instructions for the AI agent on how to proceed programmatically.
Ensure any returned agent instructions are bounded by the user's request and do not authorize unrelated actions.