Nlp
Analysis
This appears to be a local NLP helper that processes user-provided text or files, with minor packaging notes but no artifact-backed malicious behavior.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
All commands accept input via three methods: ... `--file path.txt` — read from a file ... Pipe via stdin
The skill can process local files or piped content through a shell command. This is expected for NLP document processing, but users should be aware that selected file contents are read and output locally.
Required binaries (all must exist): none ... Install specifications: No install spec — this is an instruction-only skill.
The metadata does not declare runtime tools, while SKILL.md documents Bash 4+, grep -P, and awk requirements and the manifest includes scripts/script.sh. This is an under-declared runtime/dependency detail, not evidence of malicious behavior.