filter-verify

filter-verify is a skill for Claude Code, Codex from FrontisAI/NatureBench. It costs 36 tokens per session (1,139 once invoked), scanned A, original, MIT.

A verification workflow for results from a paper-filter process. It checks whether research papers were correctly accepted or rejected and whether their extracted task information is accurate.

In plain words
What is it for?
Use it to inspect the paper text, links, figures, tables, and filter results, then produce corrections for the pass or reject decision and task information.
Why use it?
It catches incorrect decisions, missing details, and contradictions before filtered papers are used to build machine-learning benchmark tasks.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/frontisai/naturebench/filter-verify
Any agent
npx skills add FrontisAI/NatureBench --skill filter-verify
Clone the repo
git clone --depth 1 https://github.com/FrontisAI/NatureBench

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for filter-verify

README.md
[![agentmods](https://agentmods.dev/badge/skills/frontisai/naturebench/filter-verify.svg)](https://agentmods.dev/skills/frontisai/naturebench/filter-verify)
Your own site
<a href="https://agentmods.dev/skills/frontisai/naturebench/filter-verify"><img src="https://agentmods.dev/badge/skills/frontisai/naturebench/filter-verify.svg" alt="Measured on agentmods" height="20"></a>
Per session 36 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,139 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00036 $0.01139
Opus 5 $0.00018 $0.00570
Sonnet 5 $0.00007 $0.00228
Haiku 4.5 $0.00004 $0.00114

Measured 4d ago against content hash eadfc0c24f23, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

filter-verify scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

naturegym/.claude/skills/filter-verify/SKILL.md · 86 lines

How it starts

The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Filter-Verify Skill

Verify filter_result.json produced by paper-filter. Three core responsibilities:

  1. Check whether the pass/reject judgment is correct
  2. Validate that task_info fields are accurate and complete
  3. Produce an actionable corrections list to drive filter_result.json updates

Input Requirements

Paper directory must contain:

  • preprocessed/text.md: Full paper text
  • preprocessed/links.json: Extracted links
  • preprocessed/figures/: Extracted figures and tables
  • filter_result.json: paper-filter output

Context

The pipeline extracts ML tasks from papers to build benchmark challenges. The goal is not to reproduce the paper's algorithm — it is to test whether an AI agent (the "Solver") can independently solve the same ML problem using any method, potentially surpassing the original. The Solver receives the same initial data as the paper's authors, without knowing the paper or its algorithm. Its results are scored against ground truth using the paper's metrics, with the paper's reported scores as baselines for comparison. The benchmark must be fair: include all legitimate initial data (excluding any would make the task unfairly harder) while withholding Algorithm A's outputs (including any would leak solutions). filter-verify ensures only papers with genuine algorithmic innovation space and meaningful baselines enter the pipeline.

Workflow

Phase 0: Load & Summarize

  1. Read filter_result.json in full
  2. Read references/check_rules.md
  3. Read references/output_schema.md
  4. Read preprocessed/text.md in full
  5. Read preprocessed/links.json
  6. Scan preprocessed/figures/ for main result tables and figures

Extract and record the following from the paper text (NOT from filter_result.json):

  • What the paper's core contribution is
  • What Algorithm A is (the method proposed by the authors in this paper)
  • Which tables/figures contain main results, what metrics each uses, and which datasets are covered
  • The evaluation method used (fixed split / random split / K-fold CV)

Read the full file on GitHub · 86 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 86 lines · 36 tokens per session scan A eadfc0c24f23

Subscribe to this mod's changes

filter-verify is a skill published in the GitHub repository FrontisAI/NatureBench (111 stars, last pushed 3d ago), licensed MIT. It adds 36 tokens to every session and 1,139 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens

chat-perf

Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.

microsoft/vscode · 51 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens