build-guide

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from various project files including README.md, pin_config.h, and sdkconfig.defaults. This creates a potential surface for indirect prompt injection if those files contain malicious instructions.\n
  • Ingestion points: Local project files at the provided path $1 (e.g., main/pin_config.h, partitions.csv, justfile).\n
  • Boundary markers: None specified; the agent is directed to treat these files as authoritative sources.\n
  • Capability inventory: The skill utilizes Bash for command execution and Write/Edit for file modification.\n
  • Sanitization: There are no explicit sanitization steps for the data extracted from the project files before it is processed by the agent or written to the output files.\n- [EXTERNAL_DOWNLOADS]: The skill provides instructions to install the typst-cli compiler using cargo install from the official and well-known crates.io registry. This is a standard dependency management practice for the required build environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 09:35 AM
Security Audit — agent-trust-hub — build-guide