dh-cli-usage

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for the agent to resolve and execute local CLI tools using uv run. These tools, including cli.py and various management scripts like run_bounded.py, are used for DH plan operations and diagnostic checks. The commands are scoped to the skill's installation directory or the active workspace.
  • [EXTERNAL_DOWNLOADS]: The documentation for MCP connection diagnostics involves using uvx to download and run the fastmcp-slim package. This is a standard procedure for debugging Model Context Protocol server issues and uses a well-known toolchain.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes an interface for appending tasks using YAML data provided via stdin.
  • Ingestion points: External data is ingested through the plan append-task --stdin command documented in references/command-reference.md.
  • Boundary markers: The instructions do not define specific delimiters for this input, relying on the YAML structure.
  • Capability inventory: The skill has the capability to execute shell commands (uv run, uvx) and perform repository operations.
  • Sanitization: There is no explicit sanitization described in the skill instructions; it is assumed the underlying CLI validates the input schema.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 12:46 PM