An agent skill that turns LLM agents into rigorous literature assistants, finding supporting literature and auditing citations for existence, correctness, and claim support.

Stars

22

7-day growth

No data

Forks

0

Open issues

0

License

MIT

Last updated

2026-07-30

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It directly addresses the problem of hallucinated and misattributed citations in academic writing, using multiple API sources with automatic fallbacks and honesty mechanisms that clearly label uncertainty.

Who it is for

  • Graduate students writing theses or dissertations
  • Researchers who need to verify citation accuracy
  • LLM agent users seeking rigorous literature support
  • Academic editors or reviewers auditing references

Use cases

  • Audit all citations in a draft for existence, bibliography accuracy, and claim support
  • Search for supporting literature given a paragraph or topic
  • Write a literature-grounded article with citations that are self-audited
  • Quickly verify a single citation's existence and bibliographic details

Strengths

  • Uses three API sources (Semantic Scholar, OpenAlex, arXiv) with automatic fallbacks and rate limiting
  • Implements isolation firewalls (evidence, role, injection isolation) to prevent LLM hallucination
  • Provides cost tiers (quick/default/thorough) matching verification intensity to risk
  • Works with no API keys required and is pure Python 3.8+ standard library

Considerations

  • Default judgments are abstract-based; full-text verification depends on open-access availability
  • Coverage of Chinese-language literature and pre-2010 books is limited, with honest flags
  • Industry proceedings (e.g., IPC/SMTA) are near-impossible to verify; tool flags as 'not found' but never fabricates

README quick start

Quick start

Description

Literature-fetching & citation-auditing skill for LLM agents (Claude/Codex) — 讓 AI 引用經得起追問:找文獻、查核引用真偽與支持度、文獻支撐寫作,防幻覺引用。

Related repositories

Similar projects matched by category, topics, and programming language.

makecindy
Featured
makecindy GitHub avatar

cindy

Cindy is an open-source AI agent that runs locally on your machine, integrates multiple AI harnesses and models, and provides memory, skills, and automation to perform real work in your projects and apps.

AI & Machine LearningLarge Language Models
958
uzairansaruzi
Featured
uzairansaruzi GitHub avatar

hermex

Hermex is a native SwiftUI iPhone app that lets you control a self-hosted Hermes AI agent directly from your phone, with no subscriptions, tracking, or third-party relay.

AI & Machine LearningLarge Language Models
941
yetone
Featured
yetone GitHub avatar

kill-ai-slop

Kill AI Slop is a multilingual field guide and agent skill that catalogues 33 common AI-generated design tells with interactive before-after demos, and provides a tool to scan and fix web projects.

AI & Machine LearningAI Agents
805