The Agent Skills Directory

[DYNAMIC_EXECUTION]: The script coach.py dynamically loads a Python module from a relative path (../pipeline/deals.py) using importlib.util.spec_from_file_location. This mechanism allows the execution of code from a file located outside the skill's specific directory.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes live captions from external meeting sources via a local file (closer/calls/live.txt), creating a potential attack surface for indirect prompt injection if malicious instructions are embedded in the captions.
Ingestion points: closer/calls/live.txt (transcript data).
Boundary markers: The skill does not use specific delimiters or instructions to ignore potential commands within the transcript text.
Capability inventory: The script performs regex analysis and prints output; the agent is also instructed to read browser content and write to the local file system.
Sanitization: No sanitization is performed on the transcript text prior to processing beyond standard regular expression matching.

sales-coach