Project management
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from a local SQLite database and multiple Markdown files which are used to guide agent behavior.
- Ingestion points: The agent reads the
promptfield from thetaskstable in/home/box/agent-data/grok-ship/factory.db, as well as content from/home/box/agent-data/grok-ship/reports/,/home/box/agent-data/grok-ship/pack/GROK_BOT_TRIAGE.md, and/home/box/agent-data/grok-ship/pack/GROK_BOT_CREWMATE.md. - Boundary markers: There are no instructions defining delimiters or "ignore embedded instructions" warnings for the ingested data.
- Capability inventory: The skill utilizes SQL command execution for database management and performs filesystem writes to specific data directories.
- Sanitization: No sanitization, validation, or escaping of external content is specified before the content is interpolated into the agent's workflow.
- [COMMAND_EXECUTION]: The skill requires the agent to execute SQL DDL and DML commands to initialize and maintain the project backlog database.
Audit Metadata