The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions (found in SKILL-ENHANCED.md and QUICK-REFERENCE.md) mandate the use of npx claude-flow@alpha. This triggers the download and execution of a third-party package from the NPM registry that is not associated with a trusted organization or well-known service.
[COMMAND_EXECUTION]: The orchestration workflow is heavily dependent on the execution of shell commands. This includes session management via npx, environment setup using pip, and the execution of local Python scripts (validate_skill.py and package_skill.py) that perform recursive file system operations.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted user data (the skill request description) into a 7-phase agent chain without appropriate boundary markers.
Ingestion points: Untrusted data enters the agent context in Phase 1 of SKILL-ENHANCED.md.
Boundary markers: Absent; the instructions do not specify delimiters to separate user input from system-level agent instructions.
Capability inventory: The skill possesses significant capabilities, including the ability to execute shell commands (npx), perform memory operations, and write files to the local system.
Sanitization: No validation or sanitization of the user-provided request is performed before it is passed to the specialized agents.

skill-forge