The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a Python-based CLI (nl.py) to manage interactions with the Narrative Lion API. This allows the agent to perform structured operations like search, notes create, and shot-update by executing the local script with specific arguments.
[EXTERNAL_DOWNLOADS]: The skill contains commands (download, download-shot) that fetch binary content from the narrativelion.com API and save it to the local filesystem. This is used for syncing project assets like frames and audio files to a local studio directory.
[DATA_EXFILTRATION]: The skill includes an upload mechanism (upload, upload-roll) that reads local files and sends them to the Narrative Lion service. While this is the intended functionality for video production, it involves the transmission of local data to an external server.
[PROMPT_INJECTION]: The skill retrieves and displays user-generated content (notes, insights, and search results) from the Narrative Lion service. This processed data serves as an ingestion point for untrusted content that could influence agent behavior.
Ingestion points: Data retrieved from search, notes get, and insights API calls in scripts/commands/.
Boundary markers: API outputs are presented directly to the agent without special delimiters or instructions to ignore embedded commands.
Capability inventory: The skill possesses capabilities for network communication (API), file system reads (uploads), and file system writes (downloads).
Sanitization: Content is handled via standard JSON parsing, but no additional prompt-level sanitization or filtering is applied to the retrieved text.

narrative-lion