The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection by fetching untrusted external data such as call transcripts and logs from the Bolna API.
Ingestion points: The fetch_executions_paginated.py script and curl examples in SKILL.md retrieve call data including transcripts, logs, and metadata from api.bolna.ai.
Boundary markers: The skill does not implement specific delimiters or warnings to demarcate the untrusted call content when it is passed to the agent context.
Capability inventory: The skill is restricted to read-only operations and printing data to standard output; it does not have capabilities for file system modification, shell command execution, or outbound network calls to third-party domains in any of its scripts.
Sanitization: No sanitization or filtering is performed on the transcripts or extracted data retrieved from the API.
[SAFE]: The skill interacts exclusively with the official vendor domain api.bolna.ai and uses standard Python library modules. Authentication is handled securely through the BOLNA_API_KEY environment variable, following security best practices.

get-executions