sdd-checar-ambiente
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes various git and system commands (such as git status, git ls-files, and npm config) through a local bash script to perform its diagnostic checks.
- [EXTERNAL_DOWNLOADS]: Performs an anonymous metadata lookup via the official GitHub API (api.github.com) to determine repository visibility. This targets a well-known service and does not involve downloading executable code.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from the local environment which could potentially contain malicious instructions.
- Ingestion points: Reads file names, git metadata, and configuration files including package.json, composer.json, and requirements.txt.
- Boundary markers: Skill instructions explicitly command the agent to return the diagnostic output exactly as generated, reducing the risk of the agent following instructions embedded in the analyzed data.
- Capability inventory: Accesses git, bash, and curl for environment verification.
- Sanitization: While the script formats data into a report, it does not explicitly sanitize the content of file names or configuration values before outputting them to the agent context.
Audit Metadata