editorial-theme
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE]: The skill references a sensitive credential path at
~/.build-host/credentials.json. This is used for authenticating with the build.host API. The instructions include explicit safety rules prohibiting the printing of API keys, tokens, or environment variables to logs or chat. - [COMMAND_EXECUTION]: The skill is designed to perform local repository edits and initiate project deployments. These operations are intended to be performed through the build.host API using existing user credentials.
- [INDIRECT_PROMPT_INJECTION]: The operating procedure involves reading existing project UI and constraints. This creates a surface where the agent processes untrusted data (the project source) which could contain hidden instructions. The skill mitigates this risk by defining a narrow scope of operation (applying CSS/style tokens) and requiring verification of the results.
- [EXTERNAL_DOWNLOADS]: The skill describes performing network operations to the build.host API (
POST /api/projects/:uuid/deploy) and performing live HTTP checks on returned URLs. These operations are consistent with the skill's stated purpose as a deployment and design tool for the build.host platform.
Audit Metadata