The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of the npx command to execute the @talesofai/neta-skills CLI tool for all operations, including image/video generation, asset uploads, and account management.
[EXTERNAL_DOWNLOADS]: The skill fetches the latest version of the @talesofai/neta-skills package from the NPM registry during execution via npx -y.
[DATA_EXFILTRATION]: The upload command (documented in references/media-upload.md) provides a mechanism to read local files from disk or fetch remote URLs and upload them to the Neta API.
Capability: While intended for media files, this capability represents a potential vector for data exposure if sensitive file paths (e.g., .env, ~/.ssh/id_rsa) are provided to the agent. However, the documentation specifies that the CLI performs magic-byte validation to ensure only media formats are accepted.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes external data.
Ingestion points: Data enters the context via the read_collection command output in references/collection-remix.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the reference documents when processing collection data.
Capability inventory: The skill possesses capabilities to execute CLI commands, read local files, and perform network operations.
Sanitization: No specific sanitization or validation of the fetched collection content is described in the prompt logic.

neta-creative