Back to skill

Security audit

aispeech-ainote

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent office notes and todos API skill, but it needs Review because it stores a long-lived token globally and exposes delete/write actions without clear confirmation safeguards.

Install only if you trust the Aispeech/OpenClaw integration and are comfortable granting the selected note, todo, label, knowledge, and hotword scopes. Treat AIWORK_AUTH_TOKEN as sensitive, restrict access to ~/.openclaw/.env, and require the agent to confirm exact items before any delete or broad write action.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (12)

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 23)May include surrounding context.

md
envVars:
      - name: AIWORK_AUTH_TOKEN
        required: true
        description: "思必驰开放平台安装 Skill 后下发的长期授权 token。按 OpenClaw 规范持久化到 ~/.openclaw/.env"
      - name: AIWORK_BASE_URL
        required: false
        description: "业务接口 Base URL,默认 https://aiworks.cn。"

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

The skill explicitly instructs persistence of a long-lived bearer token into a global plaintext dotenv file under the user's home directory. While this may be operationally convenient, storing durable API credentials in a broadly reused global file increases the blast radius if the host, logs, backups, or other local processes can access that file.

Content

Scanner excerpt · SKILL.md (reported line 37)May include surrounding context.

  1. 为避免网关或主机重启后丢失 authToken,首次安装时,必须把 authToken 持久化到 OpenClaw 全局运行时 dotenv 文件(路径~/.openclaw/.env,不存在则新建):
dotenv
# ~/.openclaw/.env
AIWORK_AUTH_TOKEN=你的授权凭证
AIWORK_BASE_URL=https://aiworks.cn

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
91% confidence
Finding

The documented DELETE /todo/{todoUid} operation enables state-changing deletion of user data, but the skill text does not define confirmation, authorization re-checks, or safeguards against acting on ambiguous identifiers. In an agent workflow, a mistaken match or prompt-influenced parameter choice could delete the wrong todo item.

Content

Scanner excerpt · SKILL.md (reported line 80)May include surrounding context.

md
| 查待办列表        | `GET /todo/list`              | `TODO_READ`      |
| 新建待办         | `POST /todo`                  | `TODO_WRITE`     |
| 更新待办         | `PUT /todo/{todoUid}`         | `TODO_WRITE`     |
| 删除待办         | `DELETE /todo/{todoUid}`      | `TODO_WRITE`     |
| 查用户标签知识库     | `GET /label/knowledge/user`   | `LABEL_READ`     |
| 同步用户标签知识库    | `POST /label/knowledge/user/sync` | `LABEL_WRITE` |
| 查笔记标签        | `GET /label/knowledge/note`   | `LABEL_READ`     |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
84% confidence
Finding

The DELETE /hotword/delete endpoint is a destructive write action, and the skill does not describe any safety controls around selection of the hotword to remove. Without confirmation and strict parameter validation, an agent could remove the wrong customization data or be induced to make unintended changes.

Content

Scanner excerpt · SKILL.md (reported line 89)May include surrounding context.

md
| 搜索待办个人数据库      | `POST /database/todo/search` | `KNOWLEDGE_READ` |
| 查询热词词库       | `GET /hotword/content`         | `HOTWORD_READ`   |
| 新增热词         | `POST /hotword/add`            | `HOTWORD_WRITE`  |
| 删除热词         | `DELETE /hotword/delete`       | `HOTWORD_WRITE`  |
| 修改热词         | `PUT /hotword/update`          | `HOTWORD_WRITE`  |
| 标签关联查询       | `GET /label/knowledge/relations` | `LABEL_READ`   |
| 绑定笔记到标签知识库   | `POST /label/knowledge/note/bind` | `LABEL_WRITE` |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
93% confidence
Finding

The DELETE /note/{noteUid} endpoint allows deletion of user notes, which are likely sensitive and valuable records. Even though the description says notes are moved to a recycle bin, invoking destructive note actions without confirmation, target preview, and anti-ambiguity checks creates a substantial risk of unintended data loss or privacy-impacting misuse.

Content

Scanner excerpt · SKILL.md (reported line 94)May include surrounding context.

md
| 标签关联查询       | `GET /label/knowledge/relations` | `LABEL_READ`   |
| 绑定笔记到标签知识库   | `POST /label/knowledge/note/bind` | `LABEL_WRITE` |
| 绑定待办到标签知识库   | `POST /label/knowledge/todo/bind` | `LABEL_WRITE` |
| 删除笔记(入回收站)   | `DELETE /note/{noteUid}`       | `NOTE_WRITE`     |
| 新建笔记分组       | `POST /note/group/create`      | `NOTE_WRITE`     |
| 移动笔记分组       | `PUT /note/group/move`         | `NOTE_WRITE`     |
| 删除笔记分组(入回收站) | `DELETE /note/group/{groupUid}`| `NOTE_WRITE`     |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
90% confidence
Finding

The DELETE /note/group/{groupUid} operation can remove an organizational container for notes, potentially affecting many records indirectly from the user's perspective. The skill lacks procedural safeguards such as confirmation, scope preview, or warning about consequences, making accidental or overbroad destructive actions more likely in an autonomous agent context.

Content

Scanner excerpt · SKILL.md (reported line 97)May include surrounding context.

md
| 删除笔记(入回收站)   | `DELETE /note/{noteUid}`       | `NOTE_WRITE`     |
| 新建笔记分组       | `POST /note/group/create`      | `NOTE_WRITE`     |
| 移动笔记分组       | `PUT /note/group/move`         | `NOTE_WRITE`     |
| 删除笔记分组(入回收站) | `DELETE /note/group/{groupUid}`| `NOTE_WRITE`     |

## 参数与返回约定

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill exposes multiple destructive operations such as deleting notes, note groups, todos, and hotwords, but the documentation does not require explicit confirmation, user-warning language, or safety checks before invoking them. In an agent setting, this increases the chance of accidental or overly broad deletion when a user's request is ambiguous or when identifiers are inferred incorrectly.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The API reference exposes destructive delete operations for notes, todos, and groups, but provides no guidance that callers should obtain explicit user confirmation before performing irreversible or hard-to-reverse actions. In an agent skill context, this increases the chance that an LLM-driven agent could delete user data based on ambiguous prompts, misclassification, or prompt injection, causing unintended data loss.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The document specifies a default timezone of UTC+8 for unspecified time strings, which can cause the agent to interpret user-provided dates and times incorrectly when the user is in a different locale or has not consented to that assumption. In scheduling and search flows, this can lead to querying the wrong records or creating/updating todos at unintended times.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

The description includes Chinese-only text ("思必驰办公产品skill") without indicating any user language choice or that the skill is region-specific. This can violate language/locale policy when a skill appears to require a specific language or locale by default.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

Nearly all user-facing documentation in this file is presented only in Chinese, with no indication that the user may choose another language or that the locale restriction is intentional and justified. The policy calls for flagging language or locale constraints when they are imposed without user opt-in.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

This markdown file states that the skill consumes the existing AIWORK_AUTH_TOKEN and that the token is provisioned by the platform, but it does not include any warning or guidance about the sensitivity of that credential. Under the markdown-specific warning rule, documentation should disclose behaviors that may affect privacy or system integrity when credentials are involved.

Content

No source excerpt is available for this finding.