skills/dfinity/icskills/static-site/Gen Agent Trust Hub

static-site

Pass

Audited by Gen Agent Trust Hub on Aug 8, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill allows for the execution of build and presync scripts defined in the icp.yaml configuration file. These scripts are intended for standard frontend build processes (e.g., npm run build) before the assets are synchronized to the canister.
  • [EXTERNAL_DOWNLOADS]: The documentation references the installation of CLI tools and the download of synchronization plugins from official registries and the vendor's GitHub repository. These sources are trusted and necessary for the skill's functionality.
  • [COMMAND_EXECUTION]: The instructions guide the agent in using icp CLI commands to manage canisters, including access gating and principal authorization. The skill explicitly warns against granting full controller permissions to deployment principals, recommending the more restricted authorize method instead.
  • [DATA_EXFILTRATION]: Project-specific canister identifiers are stored in local mapping files to manage environment-specific deployments. No unauthorized network exfiltration patterns were detected.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes configuration files like icp.yaml which may contain user-defined shell commands.
  • Ingestion points: icp.yaml configuration file.
  • Boundary markers: None identified.
  • Capability inventory: Shell command execution via icp deploy (build and presync steps).
  • Sanitization: None identified; relies on the integrity of the project configuration.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 8, 2026, 04:36 AM
Security Audit — agent-trust-hub — static-site