添加QQBot账号

ReviewAudited by ClawScan on May 12, 2026.

Overview

This instruction-only skill is coherent for adding QQ Bot accounts, but it involves bot credentials, live gateway configuration changes, and optional persistent agent workspaces that users should review.

Safe to consider if you intend to add a QQ Bot to OpenClaw Gateway. Before using it, back up ~/.openclaw/openclaw.json, protect any appId/clientSecret values, prefer SecretRef or the credentials store for production, and only create a separate agent workspace if you really want independent memory and persona for that bot.

Findings (3)

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.

Medium

#ASI03: Identity and Privilege Abuse

What this means

Anyone who can read the OpenClaw config may be able to obtain the QQ Bot secret and act as that bot.

Why it was flagged

The skill explicitly handles QQ Bot app secrets and may store them in plaintext configuration. This is expected for adding a bot account and is disclosed, but the credential exposure is important.

Skill content

- **Credentials are in plaintext** in `openclaw.json` by default. For production, consider SecretRef (env var or file-backed).

Recommendation

Prefer the credential store or SecretRef for production, restrict file permissions on openclaw.json, and avoid pasting real secrets into shared chats or logs.

Low

#ASI08: Cascading Failures

What this means

A malformed or unintended config change could disrupt the active Gateway or route bot messages differently right away.

Why it was flagged

Saving the config can immediately change behavior of the running Gateway. This is purpose-aligned, but mistakes in the config could affect active bot routing without a separate deployment step.

Skill content

5. **Save** — Gateway hot-reloads automatically; no restart needed

Recommendation

Back up openclaw.json before editing, validate the JSON carefully, and make changes during a safe maintenance window if the Gateway is in production.

Low

#ASI06: Memory and Context Poisoning

What this means

If enabled, the new bot may retain separate context and behavior that persists across future interactions.

Why it was flagged

The optional multi-agent path creates persistent workspace/persona files and separate memory for the new bot. The skill says to ask the user first, so this is disclosed and user-directed.

Skill content

如果用户希望新 bot 有**独立的人格和记忆**（多 agent 模式）... 在该工作区中放置 `SOUL.md`、`USER.md` 等文件定义其人格。

Recommendation

Only enable a separate agent when needed, review the workspace files, and avoid placing sensitive or untrusted instructions in SOUL.md or USER.md.