Seventeen open tools split integrity checking into five specialties — none covers more than its own lane

Asked:

“What open-source tools detect statistical errors, reporting inconsistencies, citation problems, scientific image manipulation, or reproducibility gaps, and what specific checks does each tool perform?”

A researched shortlist — not an exhaustive census — of 17 open-source or source-available tools: one dedicated statistical-reporting checker, four manuscript-consistency linters, four citation auditors, four image-integrity screeners, and four reproducibility pipelines, drawn from nine independent source hosts (one row per URL). Several of the 2025–2026 systems are early-stage prototypes; all of them flag anomalies for human review, not misconduct.

What each tool actually checks, grouped by specialty

Bar length = number of distinct documented checks per tool (semicolon-separated items in each tool's own documentation). Hover or tap a bar to read the exact checks. Colour = category.

SciPeerAI documents the longest check list — 13 distinct statistical audits including GRIM, SPRITE, Benford's Law granularity analysis and p-curve shape — yet its own row lists no scope caveat, unlike every other tool. github.com

Statcheck, the one dedicated statistical-reporting tool, recomputes p-values from APA-style test statistics but finds only about 60% of significance tests in typical psychology journals — tables and non-APA reporting escape it. statcheck.io

CiteStamp is the nuance case: it is backed by a CC0 public citation graph, mirrored and forkable, but the checker itself runs client-side with paid Pro features — open data rather than a classic open-source repository. citestamp.com

Reproducibility tools carry the heaviest machinery and the frankest disclaimers: AI4Reproducibility is alpha with uncharacterised accuracy, RECAP could not process 13 of 168 papers, and PaperRepro admits it can misscore substantial deviations. github.com

All 17 tools, checks and caveats

Tool	Specific checks	Scope / caveat	Source

Method: researched shortlist of 17 open-source or source-available scientific-integrity tools across five categories, one row per tool from nine independent source hosts (no URL supplies more than one row); compiled 2025–2026, several entries are early-stage prototypes. Bar values count semicolon-separated checks in each tool's own documentation; long check and caveat texts are truncated in the table for space and shown in full on the bar tooltips where cut. These systems flag anomalies, not misconduct — database coverage, document parsing, thresholds, environment complexity and LLM behaviour can produce false positives and negatives; human review remains necessary.

This report was generated automatically by Keenable SELECT at a user's request, from publicly available web sources linked herein. Keenable does not review, verify, or endorse its contents and makes no representation as to accuracy, completeness, or timeliness; AI-based extraction may contain errors. Nothing in this report is investment, legal, financial, or other professional advice. All trademarks and referenced content remain the property of their respective owners; no affiliation or endorsement is implied. To report an error, rights concern, or request removal: legal@keenable.ai.