scout-manual-workflow
Pass
Audited by Gen Agent Trust Hub on May 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands, specifically
curlfor API interactions andgitfor repository management, to automate task-related actions within the developer environment. - [DATA_EXFILTRATION]: The skill performs network requests to fetch and update task data. These requests are directed to a user-provided endpoint (
SCOUT_URL). The instructions explicitly mandate safe handling of API keys to prevent accidental exposure. - [PROMPT_INJECTION]: The skill is subject to potential indirect prompt injection as it processes data from an external source.
- Ingestion points: Task metadata, descriptions, and comments retrieved via the Scout API endpoints in
SKILL.md(e.g.,/api/items/get). - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the task ingestion logic.
- Capability inventory: The agent can execute shell commands (
curl,git) and perform file system modifications within the local repository. - Sanitization: No sanitization or validation of the task content is specified before the agent processes the information.
Audit Metadata