The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external user data such as company goals and metrics to generate a markdown document, creating a potential ingestion surface for malicious instructions.\n
Ingestion points: Data provided by the user for strategic goals, team functions, and baseline metrics as defined in references/inputs.md.\n
Boundary markers: The workflow does not specify the use of delimiters or explicit "ignore" instructions to separate user-provided content from the agent's core logic.\n
Capability inventory: The skill is granted Read, Write, Glob, and Grep tools, which allows it to read existing files and write the generated okr-plan.md file to the workspace.\n
Sanitization: There are no documented procedures for validating, escaping, or filtering user-supplied text before it is processed by the model or written to the filesystem.

okr-generator