The Agent Skills Directory

[SAFE]: The skill instructions and bundled data files do not contain any prompt injections, obfuscated code, or unauthorized network operations. The behavior described in the instructions is consistent with the primary purpose of architectural code calculation.
[INDIRECT_PROMPT_INJECTION]: The skill maintains an attack surface for indirect prompt injection due to its integration with external project files.
Ingestion points: The skill reads program.json and occupancy.json from the current working directory to populate room schedules and track calculation state.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the content of these files as untrusted or to ignore embedded instructions.
Capability inventory: The skill possesses the capability to read from and write to the local file system (creating .md, .csv, and .json files).
Sanitization: While no explicit sanitization is mentioned, the skill instructions require the agent to map data into a fixed schema of IBC use types, which limits the potential for execution of injected instructions.

occupancy-calculator