Back to skill

Security audit

语雀 Skill

Security checks for vulnerabilities and agentic risk

Overview

This Yuque skill is coherent for document automation, but it grants broad account-level document access with unsafe defaults that could expose tokens or unintentionally modify/delete content.

Review this skill carefully before installing. Use a least-privilege Yuque token, keep the default official API URL unless you trust a private deployment, back up important repositories before TOC or batch operations, and require explicit approval before delete, raw API, import, replace, or TOC remove actions.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (4)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/yuque_client.py:41
Finding

Yuque token can be transmitted to an untrusted configurable destination

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/toc.py:122
Finding

Removing a TOC node deletes the underlying document by default

Content
View full analysis
--uuid ` without `--keep-doc`. 3. The parser leaves `args.keep_doc` as `False`. 4. The request body sets `delete_doc` to `True`. 5. Yuque removes the TOC node and deletes the underlying document. 6. The user loses content even though the intended action may only have been structural. ### Impact Assessment Exploitation or accidental triggering can permanently delete a Yuque document accessible to the configured token. Repeated ...[truncated 292 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/batch.py:102
Finding

Repository-wide regular-expression replacement writes changes by default

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/format_helpers.py:61
Finding

Markdown helpers permit HTML attribute injection and provide incomplete sanitization

Content
View full analysis
str: css: list[str] = [] if color: css.append(f"color:{color}") if size: css.append(f"font-size:{size}") ``` ```python def toc_anchor(label: str, slug: str | None = None) -> str: s = slug or re.sub(r"\s+", "-", label.strip().lower()) return f'{label}' ``` ```python _BAD_TAG = re.compile(r"<\s*(script|iframe|object|embed)[^>]*>.*?<\s*/\s*\1\s*>", re.IGNORECASE | re.DOTALL) def sanitize(body: str) -> str: return _BAD_TAG.sub("", body or "") ``` ### Technical Analysis The module generates raw HTML that may later be stored in Yuque documents: - `styled()` inserts caller-controlled `color` and `size` values directly into a `style` attribute without validation. - `toc_anchor()` inserts `slug` directly into a quoted HTML attribute and returns `label` without HTML escaping. - `sanitize()` removes only paired forms of four tag names: `script`, `iframe`, `object`, and `embed`. The sanitizer does not comprehensively handle: - Event-handler attributes such as `onerror` or `onclick`. - Attribute-termination injection. - Unsafe URI schemes. - Self-closing or malformed dangerous tags. - Other potentially active HTML elements. - Dangerous CSS values or constructs. - Encoded or parser-differential payloads. Regular expressions are not a reliable general-purpose HTML sanitization mechanism. The final exploitability depends on Yuque’s own Markdown and HTML sanitization, but the Skill’s helper contract represents `sanitize()` as a dangerous-HTML cleaning function despite providing only narrow fil ...[truncated 1312 chars]
Remediation
View remediation
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
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Findings (24)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
90% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · README.md (reported line 143)May include surrounding context.

六、卸载

bash
rm -rf ~/.openclaw/skills/yuque
unset YUQUE_TOKEN YUQUE_BASE_URL
rm -f ~/.yuque/config.json

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
90% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · README.md (reported line 143)May include surrounding context.

六、卸载

bash
rm -rf ~/.openclaw/skills/yuque
unset YUQUE_TOKEN YUQUE_BASE_URL
rm -f ~/.yuque/config.json

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
85% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · README.md (reported line 145)May include surrounding context.

bash
rm -rf ~/.openclaw/skills/yuque
unset YUQUE_TOKEN YUQUE_BASE_URL
rm -f ~/.yuque/config.json

七、许可证

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

This mismatch is security-relevant because the finding indicates the code may expose a lower-level raw API request primitive capable of arbitrary GET/POST/PUT/DELETE beyond the declared high-level use case. A skill presented as narrow Yuque document automation but actually permitting generic API calls can be abused for unintended data access, modification, or destructive operations under the user's token.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

This mismatch is security-relevant because the finding indicates the code may expose a lower-level raw API request primitive capable of arbitrary GET/POST/PUT/DELETE beyond the declared high-level use case. A skill presented as narrow Yuque document automation but actually permitting generic API calls can be abused for unintended data access, modification, or destructive operations under the user's token.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

This mismatch is security-relevant because the finding indicates the code may expose a lower-level raw API request primitive capable of arbitrary GET/POST/PUT/DELETE beyond the declared high-level use case. A skill presented as narrow Yuque document automation but actually permitting generic API calls can be abused for unintended data access, modification, or destructive operations under the user's token.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding

This mismatch is security-relevant because the finding indicates the code may expose a lower-level raw API request primitive capable of arbitrary GET/POST/PUT/DELETE beyond the declared high-level use case. A skill presented as narrow Yuque document automation but actually permitting generic API calls can be abused for unintended data access, modification, or destructive operations under the user's token.

Content

No source excerpt is available for this finding.

Vague Triggers

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill instructs automatic invocation whenever broad keywords like 'Yuque', 'knowledge base', 'document directory', or 'TOC' appear, regardless of whether the user actually requested an external action. In a skill with network, file, and credential use, such broad trigger rules can cause unnecessary exposure of tokens, accidental reads/writes, or unintended remote operations from casual discussion alone.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill documentation is presented entirely in Chinese from the title onward, with no indication that other languages are supported or that the language choice is optional. Per the policy rule, forcing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · README.md (reported line 35)May include surrounding context.

前提:先从 Releases 下载 yuque.skill,解压后得到 yuque-skill/ 目录,再执行:

bash
mkdir -p ~/.openclaw/skills/
cp -r yuque-skill ~/.openclaw/skills/yuque

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · README.md (reported line 35)May include surrounding context.

前提:先从 Releases 下载 yuque.skill,解压后得到 yuque-skill/ 目录,再执行:

bash
mkdir -p ~/.openclaw/skills/
cp -r yuque-skill ~/.openclaw/skills/yuque

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · README.md (reported line 109)May include surrounding context.

"base_url": "https://www.yuque.com/api/v2" } EOF chmod 600 ~/.yuque/config.json

text

## 三、自检

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding

The skill advertises and instructs use of capabilities that inherently access environment variables, local files, and the network, yet it declares no explicit tool scope or permission boundaries. That creates an overprivileged integration surface where the runtime may grant broader access than users or reviewers expect, increasing the blast radius of prompt injection, misuse, or implementation bugs.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The usage guidance says mentioning 'Yuque' or 'yuque' is sufficient to use the skill, without requiring context that the user wants account access or document operations. This overly broad activation logic increases the chance of privacy-invasive or destructive actions being initiated from ambiguous prompts, especially given the skill's stated support for CRUD and bulk operations.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

“用人话告诉用户做了什么”后给出的唯一示例和整份说明均默认中文语境,且文件没有说明应根据用户语言偏好切换输出语言。对于组织语言/locale 政策,若默认强制单一语言而无用户选择,可能构成自然语言策略违规。

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

This markdown reference lists create, update, and delete repository operations, including DELETE /repos/:namespace, but does not include any warning that these actions can modify or permanently remove user data. Under the markdown-specific warning criterion, skills describing behaviors that affect user data or system integrity should disclose that risk clearly.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The file describes POST, PUT, and DELETE operations for documents and later TOC mutation actions such as removeNode and delete_doc, but provides no accompanying caution about the risk of altering or deleting content. Because this is a markdown skill reference that describes data-affecting behavior, an explicit warning is expected.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The delete subcommand performs a destructive document deletion via client.delete_doc(...) but provides no confirmation prompt, warning comment/docstring, or user-facing disclosure at the point of execution. For a code file, destructive or irreversible operations should have some visible warning unless clearly disclosed elsewhere in the skill description, which is not present in this file.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

This code embeds user-facing natural-language descriptions in Chinese only, including the module docstring and command descriptions. The policy forbids forcing a specific language without user opt-in unless the locale restriction is clearly documented and justified, which is not present here.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The CLI description and help strings shown to users are fixed to Chinese, which can impose a language choice on all users. There is no indication that the user can select a locale, nor any documented justification for a Chinese-only interface.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The remove command sets delete_doc to true by default unless --keep-doc is supplied, so a user intending only to remove a TOC entry can accidentally delete the underlying Yuque document. In an automation skill that proactively operates on knowledge bases and documents, this default materially increases the risk of unintended destructive actions and irreversible content loss.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

This Python code exposes a raw CLI command that allows arbitrary POST/PUT/DELETE requests to the Yuque API, which can modify or delete remote user data. While the file documents general API usage, these destructive-capable operations have no confirmation prompt or explicit user-facing warning in the CLI help strings around the command itself.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

All natural-language guidance in this file is Chinese, and there is no indication that the skill is intentionally limited to Chinese-speaking users or that another language option is available. Per the language/locale policy, forcing a specific language without user opt-in can be a policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

This markdown file presents all guidance in Chinese and does not indicate that the language is optional, selectable, or required for a region-specific audience. Under the policy rule for language/locale constraints, forcing a single language without user opt-in can be a natural-language policy violation.

Content

No source excerpt is available for this finding.