developer-relations-kickoff
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process data from the environment, including
devrel-context.md,README.md,CONTRIBUTING.md, and git history, to automatically configure task routing. This creates an attack surface where a malicious repository could influence the agent's decision-making process. - Ingestion points: Section 1 of
SKILL.mdinstructs the agent to inventory repository files and read git logs to derive project state. - Boundary markers: The instructions lack specific delimiters or warnings for the agent to distinguish between its own logic and potentially adversarial content in the ingested files.
- Capability inventory: The skill is capable of performing file writes to the project root and executing shell commands for repository inspection.
- Sanitization: Data read from external files is processed to shape logic without explicit validation or escaping.
- [COMMAND_EXECUTION]: The skill uses local shell commands to inspect the project's development history.
- Evidence:
SKILL.md(Section 1.2) directs the agent to execute git commands to infer the stage and pace of the project from the recent log.
Audit Metadata