Resolve citations and verify claim grounding via semantic analysis.
npx skills add https://github.com/quarkloop/agent --skill service-citation
Please help me install this Agent Skill. Skill: service-citation Source: https://github.com/quarkloop/agent/tree/main/plugins/services/citation Command: npx skills add https://github.com/quarkloop/agent --skill service-citation
This Skill resolves citations and verifies grounding for claims, ensuring the accuracy of referenced sources in textual evidence.
Use the service-citation skill to verify grounding for the claim "The Quark Agent is a local operating environment for autonomous AI workspaces."
High-intent search queries and answers about installing and using this skill.
To verify grounding for claims, you can use semantic analysis to mechanically confirm that provided citation spans support selected textual claims. This ensures referenced sources are accurate and correctly grounded in evidence-based reporting.
The best way to resolve citations is to normalize selected claim text into source spans with offsets and confidence scores. This process creates normalized citation records from your selected evidence for manual verification.
Yes, you can generate citation records by normalizing selected knowledge extraction text into source spans. This creates verified citation records with offsets and confidence scores to ground your claims in the original evidence.
Coverage scoring works by summarizing citation coverage across all claims in your final answers or index writes. It generates a score reflecting how comprehensively your citations ground the extracted textual evidence.
Yes, you need a reliable agent loop and access to a context indexer to verify citation grounding. These components are required to perform the semantic analysis needed to mechanically verify that citation spans ground selected claims.
You should avoid using semantic analysis for evidence verification when you lack a reliable agent loop or indexer access. Without these dependencies, the tool cannot resolve source spans or mechanically verify that citation spans ground selected claims.
wzyxdwllExtract reusable learnings from development logs into dated .context/learnings files.
amompigSearch biomedical literature and verify citations across PubMed, Google Scholar, and arXiv.
munluckyConverts text documents into structured knowledge graphs and reusable skills.
anilcan-karaAssemble public-record evidence chains from multiple sources into normalized CSVs and findings.json.
