Run three-stage Reality Check analyses on web sources with Python and RC-CLI tools.
npx skills add https://github.com/lhl/realitycheck --skill check-lhl
Please help me install this Agent Skill. Skill: check Source: https://github.com/lhl/realitycheck/tree/main/integrations/claude/skills/check Command: npx skills add https://github.com/lhl/realitycheck --skill check-lhl
This Skill automates the end-to-end Reality Check analysis workflow for sources, enabling rapid ingestion, consistent evaluation, and provenance-backed claims registration.
Run the skill against a URL: check <url> [--domain <DOMAIN>] [--quick] [--no-register] [--continue]
High-intent search queries and answers about installing and using this skill.
Automate source evaluation by executing a single command against a URL to fetch sources, run the three-stage analysis pipeline, extract claims, and register them in a provenance-tracked database.
The three-stage analysis pipeline consists of descriptive, evaluative, and dialectical stages to systematically process sources and extract claims suitable for database import and future synthesis.
Validate results and track provenance using rc-db and rc-validate tools, which capture reasoning trails and ensure deterministic data ingestion throughout the analysis workflow.
Yes, you can run a quick reality check without registering claims by passing the --quick and --no-register flags when executing the command against your target URL.
Yes, the end-to-end source analysis workflow relies on Python scripts for database operations and validation, alongside CLI tools to orchestrate deterministic data ingestion and analysis.
Handle domain-specific sources during evidence evaluation by specifying the target domain with the --domain flag to ensure the analysis pipeline applies the appropriate contextual processing.
shige1014-devFetch web interface guidelines and audit UI files for compliance.
tkstangMerge research analysis artifacts into a comprehensive report with provenance tracking.
brunoasmExtract structured data from scientific PDFs and export to Python, R, Excel, CSV, or SQLite.
francomascareloaiTriangulate claims across academic, GitHub, and empirical sources with confidence levels.
