verdict
Pass
Audited by Gen Agent Trust Hub on Jul 18, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted data from the user via the $ARGUMENTS parameter. This data is used to construct an 'argument graph' through subsequent shell commands. There are no boundary markers or instructions provided to the agent to treat the user content as untrusted, which could allow a user to inject instructions that deviate from the skill's intended logic.\n
- Ingestion points: SKILL.md via the $ARGUMENTS parameter.\n
- Boundary markers: None identified in the prompt templates.\n
- Capability inventory: Shell command execution (uv run) documented in SKILL.md.\n
- Sanitization: No explicit sanitization or escaping of user input before command interpolation is defined.\n- [COMMAND_EXECUTION]: The workflow involves running a Python script (argument-graph.py) using uv run. User-derived content is passed directly as command-line arguments. If the user input contains shell metacharacters and the execution environment does not properly escape them, this could lead to unintended command execution.
Audit Metadata