ClawHub

Research Claim Checker

v1.0.0

检查一篇研究或分析里的结论是否被证据支撑，指出证据链断点。；use for research, claims, evidence workflows；do not use for 伪造出处, 替代同行评议.

⭐ 0· 76·0 current·0 all-time
byvx：17605205782@52yuanchangxing
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description (claim checking, evidence audit) align with included files and the provided script. The only run-time requirement is python3 and there are no environment variables, credentials, or external services requested.
ℹ
Instruction Scope
SKILL.md confines behavior to producing structured audit output and suggests running the bundled scripts. The script legitimately reads input files or directories and produces reports. Caution: when pointed at a directory it recursively reads many text file types and emits sampled content (including masked secret-like snippets), so do not run it against directories containing sensitive secrets or private data unless you intend to audit them.
✓
Install Mechanism
No install spec; this is an instruction-only skill with a local Python script. There are no downloads, package installs, or external installers in the bundle.
✓
Credentials
The skill requests no environment variables, no credentials, and requires only python3—proportionate for a local auditing/templating tool.
✓
Persistence & Privilege
always is false and the skill does not attempt to modify other skills or system-wide configuration. It writes output only when invoked with an explicit output path (or will write if default behavior used), which is normal for a report generator.
Assessment
This skill appears coherent and local-only, but follow these precautions before running: (1) review scripts/run.py yourself (it is included) to confirm behavior; (2) avoid pointing --input at home, /, or any directory containing secrets or private keys because the script will read many text files and may include snippets in its report; (3) run the smoke test in an isolated environment or with --dry-run and an example input first; (4) provide only sanitized inputs if they contain personal or sensitive data; (5) there are no network calls in the bundle, but if you modify the skill, re-check for added external endpoints.

Like a lobster shell, security has layers — review code before you run it.

latestvk9748wdsanpqa1zxas7zq4c2fx83ejma

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Runtime requirements

🔍 Clawdis
OSmacOS · Linux · Windows
Binspython3

Comments