manage-workspace

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a series of local maintenance and configuration scripts (e.g., check-dependencies.sh, generate-settings.sh) to manage the workspace environment.
  • [EXTERNAL_DOWNLOADS]: It facilitates the installation of optional Python dependencies and MCP servers via standard package managers (pip, git). These operations are handled by local scripts in the plugin root.
  • [PROMPT_INJECTION]: The skill ingests data from local directory scans and configuration files to determine the workspace state, which is a standard surface for indirect prompt injection in configuration agents.
  • Ingestion points: Plugin directory structures and .workspace-config.json.
  • Boundary markers: Not explicitly defined in the execution flow.
  • Capability inventory: File system writes, bash script execution, and package installation across multiple files.
  • Sanitization: Relies on jq for processing structured JSON data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 01:51 PM
Security Audit — agent-trust-hub — manage-workspace