csag-extraction

csag-extraction is a skill for Claude Code, Codex from fmschulz/omics-skills. It costs 38 tokens per session (2,942 once invoked), scanned A, original, MIT.

A manuscript-extraction tool that turns a research paper into a structured map of claims, contexts, evidence, and reasoning steps. It also creates grounded questions and answers from the paper.

In plain words
What is it for?
Use it to represent a manuscript in machine-readable form, validate the extracted data, and check its extraction quality.
Why use it?
It keeps extracted claims tied to the exact text that supports them and enforces consistent rules for support, contradiction, and inference.

Skill for Claude CodeCodex

Written for Claude Code and Codex: shipped in a Claude Code plugin, but also agents/openai.yaml present.

Part of the omics-skills plugin — 35 skills, 4 agents shipped together

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/fmschulz/omics-skills/csag-extraction
Any agent
npx skills add fmschulz/omics-skills --skill csag-extraction
Clone the repo
git clone --depth 1 https://github.com/fmschulz/omics-skills

Made for: Claude Code, Codex.

Or install omics-skills, the plugin that ships this one along with the rest of its 35 skills, 4 agents.

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 csag-extraction

README.md
[![agentmods](https://agentmods.dev/badge/skills/fmschulz/omics-skills/csag-extraction.svg)](https://agentmods.dev/skills/fmschulz/omics-skills/csag-extraction)
Your own site
<a href="https://agentmods.dev/skills/fmschulz/omics-skills/csag-extraction"><img src="https://agentmods.dev/badge/skills/fmschulz/omics-skills/csag-extraction.svg" alt="Measured on agentmods" height="20"></a>
Per session 38 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,942 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 5.1 $0.00038 $0.02942
Opus 5 $0.00019 $0.01471
Sonnet 5 $0.00008 $0.00588
Haiku 4.5 $0.00004 $0.00294

Measured 6d ago against content hash ff4e2140cfb6, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

csag-extraction 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 6d ago.

The scan reads SKILL.md. This mod also ships 2 executable files (scripts/csag_quality_report.py, scripts/validate_paper_extraction.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

skills/csag-extraction/SKILL.md · 264 lines

How it starts

The opening of the file, as written. The whole thing — 264 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CSAG extraction skill

Goal

Convert a manuscript into a CSAG PaperExtraction instance that is:

  • Schema-valid (LinkML: assets/csag.yaml)
  • Evidence-grounded (TextSpans for key objects)
  • Canonical (support/refute only via EvidenceLink, chains via InferenceStep)
  • Conditional (every Assertion has ≥1 Context)

Files in this skill

  • assets/csag.yaml — authoritative schema
  • assets/csag_qa_templates.yaml — QA template catalog
  • references/CSAG_PLAYBOOK.md — detailed extraction guide + edge cases

Quick Reference

Task Action
Extract paper Build one PaperExtraction per manuscript, not per search hit.
Ground claims Attach important assertions, evidence, and links to TextSpans.
Validate schema Run scripts/validate_paper_extraction.py before finalizing.
Review quality Run scripts/csag_quality_report.py --strict and resolve issues.

Non‑negotiable invariants

  1. Every Assertion MUST have ≥1 Context (schema-enforced).
  2. Support/refute polarity ONLY in EvidenceLink.
  3. Contradictions/qualification ONLY in AssertionRelation.
  4. Reasoning chains ONLY as InferenceStep.
  5. Ground important objects to TextSpans.
  6. Every Assertion MUST have normalization_status:
    • raw / partially_normalized / fully_normalized

Extraction scope

The extraction scope is the full manuscript: title, abstract, introduction, methods, results, discussion, conclusion, and supplementary material when available. If retrieval was driven by topic terms (organisms, genes, methods), do not restrict extracted assertions or evidence to sentences that mention those terms — the retrieval scope is not the extraction scope.

Instructions

Reliability pattern for model-assisted extraction

When a model is used to assist extraction, prefer a two-step workflow:

  1. Draft the scientific content first: claims, evidence snippets, evidence polarity, inferences, critiques, gaps, artifacts, datasets, and exact source quotes.
  2. Let tooling assemble the final PaperExtraction: deterministic IDs, reference fields, enum normalization, offset lookup from exact_text, and validator repair for mechanical schema-shape issues.

Read the full file on GitHub · 264 lines

Files

What ships with it

6 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. 6d ago First seen · 264 lines · 38 tokens per session scan A ff4e2140cfb6

Subscribe to this mod's changes

csag-extraction is a skill published in the GitHub repository fmschulz/omics-skills (7 stars, last pushed 11d ago), licensed MIT. It adds 38 tokens to every session and 2,942 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 2026-08-31.

Related

Other skills, from other repositories

imaging-data-commons

Query and download public cancer imaging data from NCI Imaging Data Commons. Invoke for any question about IDC collections, cancer imaging datasets, DICOM data access, radiology (CT, MR, PET) or pathology AI training sets, metadata queries, visualization, or license checks — even when the user doesn't explicitly…

K-Dense-AI/scientific-agent-skills · 75 tokens

tamarind

Access a collection of open-source molecular design and structural biology tools on the Tamarind Bio platform, via its REST API or MCP server — no local GPUs required. Tamarind bundles popular open-source models for structure prediction (AlphaFold, Boltz, Chai, ESMFold), protein, binder, and de novo design…

K-Dense-AI/scientific-agent-skills · 172 tokens

lab-hardware-cad

Design custom laboratory hardware as parametric build123d models and export fabrication-ready STEP, STL, and DXF files - microfluidic chips and molds, optomechanical mounts and breadboard adapters, cuvette and microplate holders, tube racks, animal-behavior rigs, and 3D-printed instrument fixtures. Use when a research…

K-Dense-AI/scientific-agent-skills · 106 tokens

pkpd-modeling

Pharmacokinetic and pharmacodynamic modelling and simulation - non-compartmental analysis, compartmental and population PK, PK/PD and exposure-response, TMDD, PBPK orientation, bioequivalence, allometric scaling and first-in-human dose, drug interaction prediction, and Bayesian therapeutic drug monitoring. Use when…

K-Dense-AI/scientific-agent-skills · 273 tokens

relsa-severity-assessment

Multivariate severity assessment and humane endpoint prediction for laboratory animal studies using the RELSA (RELative Severity Assessment) score and ARIMA-based foRcast forecasting. Use when combining welfare readouts — body weight or weight loss, body temperature, clinical or nesting scores, biomarkers, activity…

K-Dense-AI/scientific-agent-skills · 182 tokens

scientific-slides

Build slide decks and presentations for research talks. Use this for making PowerPoint slides, conference presentations, seminar talks, research presentations, thesis defense slides, or any scientific talk. Provides slide structure, design templates, timing guidance, and visual validation. Works with PowerPoint and…

K-Dense-AI/scientific-agent-skills · 65 tokens