pedagogy-review

Pass

Audited by Gen Agent Trust Hub on Apr 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs legitimate pedagogical reviews as described in its metadata. No malicious patterns such as obfuscation, persistence mechanisms, or privilege escalation were found.
  • [DATA_EXFILTRATION]: The skill uses Read and Write tools to process lecture slides and save reports. These operations are limited to the user-provided files and a local directory. No unauthorized network activity or sensitive file access (e.g., .ssh, .env) is present.
  • [PROMPT_INJECTION]: This skill possesses an indirect prompt injection surface. 1. Ingestion points: Lecture slides (QMD or TEX files) specified in arguments. 2. Boundary markers: Absent. 3. Capability inventory: Read, Grep, Glob, Write, and Task. 4. Sanitization: Absent. This surface is a necessary part of the skill's review function and does not pose a high risk given the intended use case.
  • [COMMAND_EXECUTION]: The Task tool is used to invoke a sub-agent for slide analysis. This usage is consistent with the skill's pedagogical purpose and does not involve the execution of arbitrary shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 16, 2026, 09:32 PM