The Agent Skills Directory

[SAFE]: The skill operates entirely within the agent's text processing environment. It contains no network operations, file system writes, or external code execution.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user data to generate negotiation strategies.
Ingestion points: Initial context is ingested via the $ARGUMENTS variable in SKILL.md.
Boundary markers: The skill does not explicitly define boundary markers for user input, relying instead on its internal workflow and output contract to structure the response.
Capability inventory: Analysis of SKILL.md and all referenced files in the references/, templates/, and examples/ directories confirms the absence of any high-risk capabilities such as network requests (curl/wget), shell execution, or file system modifications.
Sanitization: No specific input sanitization or escaping mechanisms are implemented; however, the output is strictly restricted to a defined markdown schema (Context, Decision, Analysis, Risks, Next Actions, Assumptions).

negotiation-voss-tactical-empathy