The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external files, which represents a potential surface for indirect prompt injection if those files contain untrusted content.
Ingestion points: Reads from memory/product.md, memory/customer-profiles/, and memory/learnings/ to establish product and customer context.
Boundary markers: None explicitly defined in the instructions to separate external data from system instructions.
Capability inventory: Uses file read and write operations to manage project context and record pricing models in memory/learnings/pricing-model.md.
Sanitization: No specific sanitization or validation of the input data from memory files is described.
[DATA_EXFILTRATION]: The skill manages local files within the memory/ directory. No network operations or external data transmissions were detected.
[COMMAND_EXECUTION]: There are no shell commands, subprocess calls, or system-level execution patterns in the skill instructions.

pricing