The Agent Skills Directory

[DYNAMIC_EXECUTION]: The script log.py dynamically resolves the path to and executes a Python module from a relative location (../pipeline/deals.py) using importlib.util. This technique allows the skill to run code residing outside its own directory based on a computed path.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources (emails and call notes) which could contain malicious instructions designed to manipulate the agent's behavior.
Ingestion points: Untrusted data enters the agent's context through JSON email files and external notes files via the sync, log, and call commands in log.py.
Boundary markers: The skill does not employ technical delimiters or security headers to separate untrusted external content from system instructions.
Capability inventory: The skill has the capability to write to the local filesystem (updating CRM records and saving notes) and executes logic from an adjacent skill.
Sanitization: There is no evidence of sanitization, filtering, or escaping of the ingested text before it is processed or logged by the agent.

record-keeper