build-acceleration
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent on how to install and run build optimization tools using standard package managers and build systems.
- Evidence: Shell commands for
apt-get install,brew install,cargo install,cmake, andninjaare present inSKILL.md. - Evidence: Configuration of
sccacheanddistccinvolving environment variables and daemon execution. - [INDIRECT_PROMPT_INJECTION]: The skill contains an attack surface where it ingests and processes external tool outputs and logs.
- Ingestion points:
SKILL.mdreferences processing outputs fromcmake --profiling-output,ninjatimings, and theiwyu.logfile generated during builds. - Boundary markers: The skill does not define specific delimiters or warnings to ignore malicious instructions that might be present in the build logs.
- Capability inventory: The skill utilizes subprocess execution for build commands, file writing (via
tee), and script execution (fix_include). - Sanitization: No sanitization or validation of the log content is performed before it is piped into
fix_include.
Audit Metadata