repo-engine
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and summarize content from untrusted external sources (files within the repository being scanned), which creates a surface for indirect prompt injection if those files contain malicious instructions targeting the agent.
- Ingestion points: The
read_fileandlist_dirtools are used to process arbitrary repository content as specified in theSKILL.mdworkflow. - Boundary markers: The instructions do not explicitly mandate the use of delimiters or warnings for the agent to ignore instructions found within the data it reads.
- Capability inventory: The skill is restricted to
read_fileandlist_dircapabilities, and network access is explicitly disabled in the metadata (requires_network: false). - Sanitization: There is no evidence of content sanitization or filtering before the data is processed by the agent.
- [SAFE]: The skill's architecture follows the principle of least privilege by using read-only tools and disabling network access, which mitigates the risk of data exfiltration or remote code execution.
- [SAFE]: The included utility scripts,
scripts/validate-output.shandtests/run-validation.sh, are restricted to local verification of output formats and do not involve suspicious operations, external downloads, or privilege escalation.
Audit Metadata