offensive-basic-exploitation
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data, including source code, binary files, web bundles, and request/response traces, to identify vulnerabilities.
- Ingestion points: Identified in SKILL.md under the 'Operational Workflow' where the agent is instructed to perform 'static code/binary inspection' on target software.
- Boundary markers: No specific delimiters or instructions to ignore embedded malicious content within the analyzed files were found.
- Capability inventory: The skill utilizes tools for file searching (rg), AST analysis, decompilation, and potentially executes build toolchains and test suites (e.g., npm test) as part of its remediation workflow.
- Sanitization: There is no mention of sanitization or validation of the content extracted from target files before it is processed.
- [COMMAND_EXECUTION]: The skill workflow includes the execution of commands such as 'npm test' and 'toolchain build & test suite' commands on the code being audited. If the target source code is malicious, these operations could lead to local code execution in the agent's environment.
Audit Metadata