async-effect-race-contracts
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent ingests and analyzes external source code from a user's repository, creating an attack surface for malicious instructions embedded in the analyzed data.
- Ingestion points: Source files located in
src/andapp/directories, which are accessed using therg(ripgrep) utility as specified inSKILL.md. - Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings for the agent to use when processing the code retrieved from files.
- Capability inventory: The skill relies on the agent's ability to execute
rgvia subprocess calls to perform code searches. - Sanitization: The skill lacks explicit sanitization or escaping requirements for the content retrieved from external files before it is processed by the model's logic.
- [SAFE]: The skill uses standard search utilities for local file analysis and does not attempt to access sensitive system files or credentials. All technical guidance provided aligns with established React best practices, and no patterns of obfuscation or unauthorized network exfiltration were found.
Audit Metadata