The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses natural instructional language to guide the agent in orchestration tasks. No attempts to override safety filters, extract system prompts, or bypass ethical guidelines were detected.
[DATA_EXFILTRATION]: No network operations (e.g., curl, wget, fetch) or access to sensitive file paths (e.g., .ssh, .aws, .env) are present in the documentation.
[OBFUSCATION]: No hidden content, Base64 encoding, zero-width characters, or homoglyph attacks were found within the text.
[REMOTE_CODE_EXECUTION]: The skill does not include any scripts, package manifests (package.json, requirements.txt), or commands to download and execute external code.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process musical specification data (ARR-SPEC), creating a theoretical ingestion point for untrusted data. However, as the skill possesses no capabilities for shell execution, network access, or file system modification, this surface is not exploitable.
[DYNAMIC_CONTEXT_INJECTION]: No load-time shell execution patterns (!command) were found in the SKILL.md frontmatter or body.
[NO_CODE]: The entire skill consists of Markdown reference files providing information and guidance to the agent, with no active code components.

mc-orchestration