common-protocol-enforcement
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill defines an internal quality control process to verify task completion and adherence to project-specific rules. It uses authoritative markers to emphasize workflow importance but does not attempt to bypass agent safety filters or exfiltrate data.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface for indirect prompt injection as it is designed to analyze untrusted data sources.
- Ingestion points: The 'Post-Write Self-Scan' section in
SKILL.mdinstructs the agent to read and analyze code diffs and file content. - Boundary markers: None are present in the skill instructions to delimit the ingested data from the auditing instructions.
- Capability inventory: The skill uses standard file reading and editing tools (
view_file,re-edit) to perform its verification functions. - Sanitization: No specific sanitization or validation logic for the ingested file content is defined. (Note: As the analysis of external content is the primary function of the auditing protocol, this finding is documented as a structural characteristic and does not escalate the safety verdict.)
Audit Metadata