session-palace-builder

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect prompt injection surface identified. The skill is designed to ingest and organize external session data without explicit security boundaries.
  • Ingestion points: Processes 'Conversations', 'Research', and 'Code' artifacts from the session context (documented in SKILL.md).
  • Boundary markers: Absent; the skill does not implement delimiters or instructions for the agent to ignore commands embedded in session data.
  • Capability inventory: Includes shell command execution via documented 'make' commands.
  • Sanitization: Absent; no evidence of data validation or escaping before processing session content.
  • [COMMAND_EXECUTION]: The documentation instructs the agent or user to execute 'make build', 'make clean', and 'make test' to verify functionality. This creates a command execution surface that relies on the contents of an external Makefile.
  • [COMMAND_EXECUTION]: The troubleshooting section suggests running with 'appropriate privileges' when encountering permission errors, which encourages the use of elevated permissions (e.g., sudo) without documenting the associated risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 06:51 PM