Back to portfolio

Portfolio case study

AntiRot - Research Artifact Linter

Open-source CLI that catches unsupported claims, broken citations, weak source anchors, and draft markers in AI-written research drafts.

Overview

AntiRot is a local-first review harness for Markdown research artifacts. It turns the final draft into a gateable surface by flagging unsupported claims, missing source anchors, citation mismatches, comparative hype, absolute overclaim language, and leftover draft markers before a paper, proposal, or lab note ships. The current public release adds paragraph-aware parsing, in-document references support, safer citation verification, and GitHub Actions coverage for text, JSON, Markdown, and SARIF outputs.

What It Covers

  • Catches unsupported claims, citation drift, hype language, comparative overreach, absolute claims, and leftover draft markers in Markdown drafts
  • Supports paragraph-aware source carry, footnotes, inline links, DOIs, arXiv ids, and in-document references sections
  • Runs locally with no API key and no network dependency, so it fits agent loops, proposals, and paper pipelines
  • Emits text, JSON, Markdown, and SARIF outputs for terminal use, CI gates, and GitHub-native review flows

Stack And Topics

  • Python
  • CLI
  • Markdown
  • SARIF
  • GitHub Actions
  • Research Agents

Public Signals

  • Public release: v0.2.0 GitHub release
  • Output formats: 4 text, json, markdown, sarif
  • Issue families: 8 unsupported, numeric, citation-not-found, citation-unverified, hype, comparative, absolute, draft markers
  • Runtime deps: 0 standard-library CLI

References