prepare-implementation-plan
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the 'az boards work-item show' command to retrieve data. This involves shell execution of a local CLI tool with user-provided work-item IDs.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from Azure DevOps work items and user input to drive its reasoning and file operations. 1. Ingestion points: The skill fetches data from Azure DevOps work items and user-provided descriptions. 2. Boundary markers: No delimiters or instructions are used to separate external data from system instructions. 3. Capability inventory: The agent is instructed to read local files, execute CLI commands, and write implementation plans to the filesystem. 4. Sanitization: There is no evidence of validation or sanitization of the retrieved work item content before it influences the agent's behavior.
Audit Metadata