slate-ar-finalize
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for executing local development tools and scripts to finalize research output.
- Evidence: The file references the execution of 'Codex Autoresearch' commands and the 'finalize-autoresearch.mjs' script.
- [PROMPT_INJECTION]: The skill utilizes user-provided arguments, creating a surface for indirect prompt injection.
- Ingestion points: The skill processes user input through the '$ARGUMENTS' variable (SKILL.md).
- Boundary markers: No explicit delimiters or boundary markers are defined in the instructions to isolate user input from the agent's system logic.
- Capability inventory: The skill has the capability to execute shell commands and run JavaScript scripts.
- Sanitization: No explicit sanitization or validation steps are described for the input data.
- Mitigation: The instructions include strict constraints that prevent the agent from performing mutations, such as branch creation or committing code, without explicit, turn-specific user approval (e.g., 'Do not create branches... without exact approval').
Audit Metadata