The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the user to install the unslop Python package from a community source using pipx or uv. This executes code that is not from a recognized official vendor.
[COMMAND_EXECUTION]: The skill provides instructions to run the unslop command-line utility, which executes locally on the user's machine.
[DATA_EXFILTRATION]: The documentation indicates that the tool's default mode uses the Anthropic API. This involves transmitting the text content being processed to an external network endpoint.
[PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection by processing arbitrary text through a pipeline that can include an LLM.
Ingestion points: Untrusted text is ingested via stdin and local files (e.g., draft.md).
Boundary markers: There are no boundary markers or instructions to ignore embedded commands within the processed text.
Capability inventory: The skill allows for file system modification (mv commands) and network communication (LLM API calls).
Sanitization: The instructions do not mention any sanitization or validation of the input text before it is processed.

unslop