Plagiarism Checker
v1.0.0Use when: User provides text/document and asks to check originality, detect plagiarism, assess similarity, or rewrite high-duplicate content. Triggers: "chec...
⭐ 0· 76·0 current·0 all-time
byAIpoch@aipoch-ai
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
Name/description (plagiarism detection + paraphrasing) match the SKILL.md and the included algorithm and paraphrasing documentation. Required capabilities (file I/O, local NLP algorithms, paraphrasing) are consistent with what is present in the repository; no unrelated credentials or tools are requested.
Instruction Scope
SKILL.md instructs running scripts/main.py on local text or files and explicitly states processing is local and input paths should be validated. The code shown implements segmentation, similarity metrics, and paraphrasing locally. However, the provided main.py in the prompt was truncated, so I cannot confirm whether the CLI implementation enforces the documented input-path validation (no '../' traversal), sanitizes errors, or includes any hidden behaviors (network calls, reading unrelated system files).
Install Mechanism
No install spec; only a lightweight requirements.txt (dataclasses). Being instruction-only plus a local script is low-risk from an install perspective — nothing is downloaded or installed beyond standard Python packages.
Credentials
No environment variables, credentials, or config paths are requested. This is proportionate for a local plagiarism tool that claims no external APIs.
Persistence & Privilege
Skill is not flagged as always: true and does not request persistent system presence. It appears to run as an on-demand script with no modifications to other skills or system-wide settings.
Assessment
This skill appears coherent and local-only, but before installing or running it: 1) review the full scripts/main.py (the prompt version was truncated) to confirm there are no network calls, telemetry, or attempts to read unrelated files; 2) verify input-path validation (reject '../' and absolute paths) to avoid path-traversal and accidental disclosure; 3) run it in a sandbox/workspace with non-sensitive test files first to confirm outputs and error handling; 4) check paraphrasing outputs carefully because automated rewrites can change meaning; and 5) consider adding dependency pinning and a brief security review if you will run it on confidential documents.Like a lobster shell, security has layers — review code before you run it.
latestvk973jj6mm4re4esrccb1tvp7qh83wy3z
License
MIT-0
Free to use, modify, and redistribute. No attribution required.