unslop
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the unslop package from a community repository using pipx or uv.
- [DATA_EXFILTRATION]: The tool can transmit input data to external LLM providers (specifically Anthropic) when not operating in deterministic mode, requiring a user-provided API key.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted prose from files or stdin through a CLI tool, creating a surface where malicious instructions in the input could influence agent behavior.
- Ingestion points: Content is read from stdin and local markdown files (e.g., draft.md, README.md, docs/*.md).
- Boundary markers: The instructions do not define delimiters or markers to isolate untrusted content from command logic.
- Capability inventory: The skill executes CLI commands, pipes data between processes, and performs file system writes via redirection.
- Sanitization: No input validation or content filtering is implemented for the text being processed.
Audit Metadata