The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions in SKILL.md reference a non-standard command-line utility voice-scorer for scoring and analyzing documents. While this appears to be a project-specific tool associated with the 'organvm-iv-taxis' ecosystem, the binary itself is not provided within the skill, and users should verify the source and integrity of such external tools before execution.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process prose-heavy documents (doc.md) to evaluate them against the 'Voice Constitution'. This creates a surface where malicious instructions embedded in the analyzed text could potentially influence agent behavior.
Ingestion points: Input documents provided for scoring via the voice-scorer CLI or direct agent review in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' commands when the agent processes the external document content.
Capability inventory: The agent is empowered to review, diff, and potentially edit documents based on the provided constitution, which involves significant processing of untrusted content.
Sanitization: There are no explicit requirements for sanitizing or filtering input text before it is analyzed for voice compliance.

voice-enforcement