The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill manages API credentials by reading from and writing to ~/.chanjing/credentials.json. This is used to store the application ID, secret key, and access tokens required for authentication with the Chanjing platform.
[EXTERNAL_DOWNLOADS]: Scripts such as scripts/download_result and scripts/upload_file perform network operations to transfer media files and synthesis results. These operations target the vendor's official infrastructure (*.chanjing.cc) and associated signed URLs.
[COMMAND_EXECUTION]: The authentication module scripts/_auth.py utilizes subprocess.run to invoke a helper script from the chanjing-credentials-guard skill. This is a controlled execution of a specific local file used to trigger the vendor's login workflow when credentials are missing.
[PROMPT_INJECTION]: The skill ingests untrusted data that could be leveraged for indirect prompt injection attacks.
Ingestion points: User-supplied text provided to the --text argument in scripts/create_task, as well as remote audio and background resources fetched via --wav-url or --bg-src-url.
Boundary markers: The scripts do not employ delimiters or explicit instructions to the model to ignore embedded commands within the processed data.
Capability inventory: The skill has the capability to write files to the local disk (scripts/download_result), execute local subprocesses (scripts/_auth.py), and perform network requests to external APIs.
Sanitization: No sanitization, filtering, or validation of the input text content is performed before it is transmitted to the synthesis API.

chanjing-video-compose