Back to skill

Security audit

CatchLinger Capture

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because it reports a bookmark-tag merge that can delete a source tag after updating only a limited set of bookmarks, and several executable scripts were only partially inspected.

Consider the suspicious verdict before installing. In particular, review the reported tag-merge behavior and whether it can delete a tag before every bookmark is migrated, and consider the use of GITHUB_TOKEN with GitHub requests. The scan is incomplete, so it does not establish that the partially inspected scripts are safe.

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 (1)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/tag-resolver.mjs
Finding

Tag merge performs destructive changes without an enforced confirmation gate

Content
View full analysis

Impact

An Agent invoking the helper before obtaining confirmation can replace bookmark tag assignments and delete the source tag. Printing the plan does not pause execution or obtain approval.

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (26)

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 15)May include surrounding context.

md
3. **但标签命名空间要克制。** 用户的标签空间是长期资产,碎片化(`gis` 与 `gis数据`、`springboot` 与 `spring boot`)会让检索越来越难用。**新增标签前必须先过一遍 `tag-resolver.mjs`**(见「标签归一」一节)—— 能复用既有标签就别新造。

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 34)May include surrounding context.

md
标签归一必须携带本次语言,例如 `node scripts/tag-resolver.mjs resolve "database" "performance optimization" --locale en-US`。先用目标语言提炼候选,再做归一;resolver 不负责翻译。语言参数会阻止中英文标签之间的自动复用,

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 119)May include surrounding context.

md
③.5 标签归一:node scripts/tag-resolver.mjs resolve "<所有候选标签>" ... --locale <输出语言>

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 180)May include surrounding context.

md
标签是回看和检索时的入口,不是文章关键词清单。保存前先用 `node scripts/tag-resolver.mjs list` 浏览已有词汇,优先选语义合适的较宽标签,再用 resolver 检查拼写与近义词。

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 207)May include surrounding context.

md
node scripts/tag-resolver.mjs resolve "标签1" "标签2" "标签3" --locale zh-CN

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 209)May include surrounding context.

md
node scripts/tag-resolver.mjs resolve "database" "performance optimization" --locale en-US

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 309)May include surrounding context.

md
node scripts/tag-resolver.mjs list # 紧凑列出既有标签(name×count)

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 310)May include surrounding context.

md
node scripts/tag-resolver.mjs resolve "标签" [...] # 解析候选标签 → 复用哪个既有标签

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 311)May include surrounding context.

md
node scripts/tag-resolver.mjs audit # 找出既有标签里的近重复分组

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 312)May include surrounding context.

md
node scripts/tag-resolver.mjs merge <from> <to> [--dry-run] # 把 from 合并进 to

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 57)May include surrounding context.

md
node scripts/fetch-content.mjs "<url>" ["<url2>" ...]

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 95)May include surrounding context.

md
> **配额提示**:列 star 列表可以免 Token,但**逐个抓仓库详情**(`fetch-content.mjs` 遇到 GitHub 链接会走 API)每个仓库消耗一次配额。star 数量多时建议配 `GITHUB_TOKEN`。

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 115)May include surrounding context.

md
② 取得正文:node scripts/fetch-content.mjs "<url1>" "<url2>" ...

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 65)May include surrounding context.

md
node scripts/github-stars.mjs [--limit 50] [--user <GitHub用户名>]

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 100)May include surrounding context.

md
node scripts/read-bookmarks.mjs [--browser chrome|edge|all] [--profile Default] [--limit 200] [--folder "书签栏"]

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 121)May include surrounding context.

md
④ 写入:node scripts/catchlinger-api.mjs save-many '[{...}]'(标签用归一后的)

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 247)May include surrounding context.

md
node scripts/catchlinger-api.mjs save '{"url":"https://...","title":"...","summary":"...","tags":["a","b"],"source":"skill"}'

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 253)May include surrounding context.

md
node scripts/catchlinger-api.mjs save-many '[{"url":"...","title":"...","summary":"...","tags":["..."]}, ...]'

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 301)May include surrounding context.

md
node scripts/catchlinger-api.mjs search --q 关键词 --limit 10 # 确认是否已收录

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 302)May include surrounding context.

md
node scripts/catchlinger-api.mjs stats # 总数 / 未读 / 超期

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 303)May include surrounding context.

md
node scripts/catchlinger-api.mjs tags # 既有标签(完整 JSON)

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, network.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/fetch-content.mjs (reported line 94)May include surrounding context.

js
const owner = m[1];
  const repo = m[2].replace(/\.git$/, '');
  const apiUrl = `https://api.github.com/repos/${owner}/${repo}`;

  const headers = { accept: 'application/vnd.github+json' };
  if (process.env.GITHUB_TOKEN) {

The script silently reads GITHUB_TOKEN and sends it to the GitHub API without disclosing credential use.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The script automatically reads GITHUB_TOKEN from the environment and includes it as a bearer credential in GitHub API requests. Although the introductory comment documents fetching repository information through the API, it does not disclose credential access or authenticated requests, and no comment, prompt, or user-facing log explains this behavior.

Content

No source excerpt is available for this finding.

Merge claims to migrate all bookmarks but processes at most 200 before deleting the source tag.

Medium
Category
Excessive Agency
Confidence
98% confidence
Finding

The merge operation fetches at most 200 matching bookmarks without pagination, then treats successful updates of that limited set as proof that every bookmark was migrated. For larger collections, it attempts to delete the source tag while unmigrated bookmarks remain, potentially removing their tag associations or leaving an incomplete merge depending on server-side deletion behavior.

Content

No source excerpt is available for this finding.