skills/dbosk/claude-skills/kth-data/Gen Agent Trust Hub

kth-data

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The kth_room.py script fetches data from the official KTH website (www.kth.se) to retrieve room specifications and capacities.\n- [COMMAND_EXECUTION]: The skill makes extensive use of the nytid command-line utility for managing schedules, TA rosters, and Zulip messages. It also executes a local Python script kth_room.py.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external web pages and internal file systems (AFS).\n
  • Ingestion points: Web scraping of kth.se via kth_room.py and reading course configurations from AFS/local storage via nytid.\n
  • Boundary markers: The skill does not implement specific delimiters for ingested data, but the instructions emphasize human review before taking outward-facing actions like posting to Zulip.\n
  • Capability inventory: Includes network requests (urllib.request), file system access (AFS and home directory), and command execution (nytid).\n
  • Sanitization: The scraping script uses HTML unescaping and regular expressions to extract structured data fields.\n- [DATA_EXPOSURE]: The skill references files that may contain sensitive information, such as TA rosters with personnummer and Zulip authentication files (.zuliprc). However, it includes explicit warnings to the agent not to expose sensitive personal data in messages or files.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 11:34 PM