raising-a-pull-request
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected. The skill is designed with safety gates, such as requiring explicit user approval for merges and ensuring CI reports are current and successful before proceeding.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources, which is a standard surface for indirect prompt injection attacks.
- Ingestion points: Pull request content (descriptions, comments), CI logs, and ticket state retrieved from provider APIs (SKILL.md).
- Boundary markers: Absent; the skill does not specify the use of delimiters or specific handling for external data strings.
- Capability inventory: Write operations including pull request merging and description updates (SKILL.md).
- Sanitization: Absent; the skill relies on the provider's API for data retrieval without explicit sanitization steps.
- Mitigation: The risk is effectively neutralized by the strict requirement that 'Readiness is never merge authorization. Merge only after the user explicitly approves it,' ensuring a human validates the agent's intent before any destructive or permanent action is taken.
Audit Metadata