The Agent Skills Directory

[NO_CODE]: The skill is composed entirely of markdown documentation and reference materials for a specific AI agent role. It contains no executable scripts (e.g., Python, JavaScript, Shell), binaries, or dependency manifest files (such as requirements.txt or package.json).
[CREDENTIALS_UNSAFE]: The skill explicitly promotes secure development practices by instructing the user not to embed license secrets in code repositories and to use environment variables or secret stores for solver licenses (identified in references/solver_integration_and_production.md).
[INDIRECT_PROMPT_INJECTION]: The skill defines a role that processes external data, establishing a potential injection surface.
Ingestion points: The references/problem_formulation_and_data.md file describes extracting parameters and scenarios from external systems including warehouses, ERP systems, GIS, and manual files.
Boundary markers: The instructions do not specify the use of delimiters or boundary markers to encapsulate and isolate this external data from the system prompt.
Capability inventory: The skill does not include its own code, but it defines the logic for interacting with external optimization solvers and production APIs.
Sanitization: The skill includes a 'Validation rules' section in references/problem_formulation_and_data.md that mandates checking bounds, capacities, and dimensional consistency of ingested data.

operations-research-algorithm-developer