rt-venue-integrity

Warn

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions provide shell command templates that directly interpolate user-supplied placeholders into a shell environment, which can be exploited for command injection. Evidence: python3 tools/match_venues.py --title "<the venue's name>" --top 5 and grep -i "<name fragment>" shared-resources/journal-selection/venue-index.tsv in SKILL.md.- [PROMPT_INJECTION]: The skill is designed to ingest and process data from untrusted external academic venues, presenting an indirect prompt injection attack surface.
  • Ingestion points: User-provided journal names and content retrieved from external sources such as Clarivate, Scopus, and DOAJ during the verification protocol (SKILL.md).
  • Boundary markers: Absent; there are no instructions to the agent to use delimiters to isolate external data from its own logic.
  • Capability inventory: Execution of local python scripts and grep commands, as well as the ability to perform network lookups against multiple external domains (SKILL.md).
  • Sanitization: Absent; no instructions are provided to escape or validate external content before processing.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 24, 2026, 10:29 AM
Security Audit — agent-trust-hub — rt-venue-integrity