Releases·Tool Release
Open-source LLM skill 'academic-defensive-writing-auditor' released on GitHub
A new open-source LLM skill and CLI, academic-defensive-writing-auditor, launched on GitHub for detecting and rewriting defensive academic writing. It targets reviewer-facing prebuttals, disclaimer-heavy prose, caveat stacking, result excuses, and AI-style automatic summaries in research papers. It explicitly preserves necessary scientific caveats and evidence boundaries, distinguishing it from a simple hedge-word remover.
39 engagement1 sourceSun, Aug 30, 2026, 06:45 AM
AI Overview
- Ships as a reusable LLM skill plus CLI for research papers, conference submissions, rebuttal cleanup, and manuscript polishing
- Detects reviewer-facing prebuttals, disclaimer-heavy prose, unnecessary caveat stacking, result excuses, and omitted-experiment defenses
- Also flags AI-style automatic summaries in manuscripts
- Explicitly not a hedge-word remover: distinguishes necessary scientific caveats and evidence boundaries from defensive writing
- Available with English and Simplified Chinese documentation
Related
Products·launched
Game the LLM Reviewer: reword papers to pass AI peer review
An open-source tool that applies a final editing pass to papers before submission, making small meaning-preserving wording changes based on studies of LLM review preferences. It helps authors defend their research against LLM reviewer bias by packaging the findings as an agent-compatible skill.
An open-source tool that applies a final editing pass to papers before submission, making small meaning-preserving wording changes based on studies of LLM review preferences. It helps authors defend their research against LLM reviewer bias by packaging the findings as an agent-compatible skill.
124 engagement1 sourcegithubMon, Sep 21, 2026, 11:11 PM
Releases·Tool Release
Grad student releases lit-review-skill for LLM agents to fetch and audit citations
A grad student built an open-source skill that turns Claude or Codex into a rigorous literature assistant. It finds supporting papers and audits existing citations for existence, correctness, and claim support, aiming to reduce hallucinated references.
A grad student built an open-source skill that turns Claude or Codex into a rigorous literature assistant. It finds supporting papers and audits existing citations for existence, correctness, and claim support, aiming to reduce hallucinated references.
18 engagement1 sourcegithubWed, Jul 29, 2026, 10:26 AM
Releases·Tool Release
sepia releases as a portable de-AI writing skill for Claude Code, Codex, Grok Build, and Antigravity
The sepia project published a portable Agent Skill that removes AI tells from writing across coding agents including Claude Code, Codex, Grok Build, and Antigravity. It repairs narrative architecture in fiction before word choice and applies venue-matched rules to professional documents. It is based on the StoryScope paper (arXiv:2604.03136).
The sepia project published a portable Agent Skill that removes AI tells from writing across coding agents including Claude Code, Codex, Grok Build, and Antigravity. It repairs narrative architecture in fiction before word choice and applies venue-matched rules to professional documents. It is based on the StoryScope paper (arXiv:2604.03136).
23 engagement1 sourcegithubFri, Aug 28, 2026, 05:56 AM
Releases·Model Release
OpenReviewer: A Specialized LLM for Generating Critical Scientific Paper Reviews
OpenReviewer is a specialized LLM designed to generate critical scientific paper reviews. It aims to assist researchers and reviewers by providing automated, critical feedback on academic manuscripts.
OpenReviewer is a specialized LLM designed to generate critical scientific paper reviews. It aims to assist researchers and reviewers by providing automated, critical feedback on academic manuscripts.
6 engagement1 sourcehackernewsTue, Jul 28, 2026, 08:39 PM
Products·launched
Deterministic check for fabricated quotes in LLM output
Verbatimeter is a tool that deterministically verifies whether quotes in LLM-generated text are fabricated. It solves the problem of hallucinated citations by checking each quote against a known source, helping researchers and developers trust LLM outputs.
Verbatimeter is a tool that deterministically verifies whether quotes in LLM-generated text are fabricated. It solves the problem of hallucinated citations by checking each quote against a known source, helping researchers and developers trust LLM outputs.
3 engagement1 sourcehackernewsTue, Jul 14, 2026, 11:53 AM