ClawHub
Back to skill

Security audit

librarian

Security checks for vulnerabilities and agentic risk

Overview

This skill saves user-provided links or files into an Obsidian-style library and discloses its fetching and note-writing behavior.

Install this if you are comfortable with the agent fetching user-shared links and saving summaries into a local library folder. Be careful when dropping private files or sensitive URLs, since the skill may summarize and store their contents, and X/Twitter links are fetched through FxTwitter.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill’s trigger conditions are extremely broad: it activates on virtually any dropped URL/file or phrases like 'save this' and 'file this'. This increases the chance of unintended invocation and automatic ingestion of sensitive or irrelevant content, especially because the skill is designed to fetch, process, and store external material into a shared library folder.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

The skill instructs the agent to send user-supplied tweet/X identifiers to a third-party service (api.fxtwitter.com). That creates an external data transmission path outside the primary platform, which can leak user interests, accessed URLs, metadata, and possibly sensitive private or internal links if the agent misclassifies content or if transformed identifiers are derived from user input.

Content

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

md
From the URL or file, determine:
- **YouTube** — fetch page title + description; note it needs transcript for deep analysis
- **Article / blog** — fetch with web_fetch
- **Tweet / X** — use FxTwitter API: `https://api.fxtwitter.com/{username}/status/{id}`
- **PDF** — use pdf tool
- **Image** — use image tool
- **Podcast / audio** — note URL + title, flag that transcript unavailable unless provided

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 · references/media-types.md (reported line 15)May include surrounding context.

md
- **What to extract:** Full text → summarize key ideas + structure

## Tweets / X posts
- **Fetch:** FxTwitter API — `https://api.fxtwitter.com/{username}/status/{tweet_id}`
  - Extract username and tweet_id from URL: `https://x.com/username/status/123456`
  - Returns JSON with full tweet text, author, engagement
- **Thread:** Fetch each tweet in thread individually if linked

Static analysis

No suspicious patterns detected.