The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches and executes the @steipete/summarize package from the npm registry using bun x in scripts/summarize.sh.- [REMOTE_CODE_EXECUTION]: Executes third-party code downloaded at runtime via the bun x command. The skill passes user-provided arguments directly to this execution environment.- [COMMAND_EXECUTION]: Interacts with local system binaries including yt-dlp, ffmpeg, tesseract, and whisper-cli for media handling and OCR. It also supports execution of arbitrary commands defined in environment variables such as SUMMARIZE_ONNX_PARAKEET_CMD for transcription processing.- [PROMPT_INJECTION]: Vulnerable to indirect prompt injection due to the processing of untrusted external data.
Ingestion points: Content is ingested from arbitrary web URLs, YouTube video transcripts, and local files provided by the user.
Boundary markers: The skill does not implement explicit delimiters or instructions to the model to ignore potential malicious prompts embedded within the ingested data.
Capability inventory: The skill can execute shell commands through its CLI wrapper, read sensitive local configuration files (~/.summarize/config.json), and write files (extracted slides) to the local filesystem.
Sanitization: No sanitization or filtering logic is present to clean external content before it is processed by the agent.

summarize