atomize

atomize is a skill for Claude Code, Codex from nemock/company-brain. It costs 110 tokens per session (2,295 once invoked), scanned A, original, MIT.

A document-ingestion workflow for turning Markdown, text, Word documents, PDFs, interview transcripts, and screenshots into linked knowledge records. Each original document is kept as a source record, while extracted information is stored as derived records.

In plain words
What is it for?
Use it to import existing documents into a company knowledge vault, create typed records from their contents, and link those records back to their source files.
Why use it?
It preserves where each piece of extracted information came from, making the resulting knowledge easier to audit. It also checks the knowledge vault's schema before writing records.

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/nemock/company-brain/atomize
Any agent
npx skills add nemock/company-brain --skill atomize
Clone the repo
git clone --depth 1 https://github.com/nemock/company-brain

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 atomize

README.md
[![agentmods](https://agentmods.dev/badge/skills/nemock/company-brain/atomize.svg)](https://agentmods.dev/skills/nemock/company-brain/atomize)
Your own site
<a href="https://agentmods.dev/skills/nemock/company-brain/atomize"><img src="https://agentmods.dev/badge/skills/nemock/company-brain/atomize.svg" alt="Measured on agentmods" height="20"></a>
Per session 110 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,295 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 $0.00110 $0.02295
Opus 5 $0.00055 $0.01148
Sonnet 5 $0.00022 $0.00459
Haiku 4.5 $0.00011 $0.00230

Measured 4d ago against content hash 7e7ab0f91076, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

atomize 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.

skills/atomize/SKILL.md · 182 lines

How it starts

The opening of the file, as written. The whole thing — 182 lines — stays where its author put it; the contents beside it link to each section on GitHub.

atomize

This skill ingests existing documents into the vault. Where the intake skill turns conversation into nodes, atomize turns files into nodes.

The pattern is the same: every ingested document gets one immutable source node holding the raw / extracted content, and the typed nodes derived from it carry derived_from edges back to the source. The source is the audit trail; the derived nodes are the productized knowledge.

Before any input format

  1. Confirm the vault path. Default: current working directory. Reject if no _system/PROFILE.md is present.
  2. Load the active schema. Run cb describe-profile --path <vault> to know which node types are active and what fields they require.
  3. For each node type you're about to write, run cb describe-node <type> so you don't miss required fields.
  4. Confirm the user's handle for verified_by.

Universal node-writing rules and edge-writing rules are documented in the intake skill (skills/intake/SKILL.md). They apply here too.

Universal source-node convention

Every atomize session starts by writing one immutable source node:

---
id: source-<source-kind>-<short-slug>-<YYYY-MM-DD>
title: "<Original document title>"
type: source
namespace: <appropriate; e.g. competitive, regulatory-references, customer-research>
summary: "One line about what this document contributes to the graph."
auto_inject: false
applicable_when: null
confidence: 0.85
verified_at: <today>
verified_by: <handle>
staleness_signal: <a concrete trigger; null is OK for frozen historical docs>
tags: [atomized]
source_kind: <one of the 14 registered values — see below>
edges: []
related: []
source_url: <if applicable>
controlled_document: false
---

# <Original document title>

## Summary

<1-2 sentences explaining what this source contributes.>

## Original content

<For text/markdown: paste verbatim.>
<For Word/PDF: paste the extracted text from `cb extract`.>
<For image: paste the vision-extracted text and reference the image file in `_attachments/`.>

Read the full file on GitHub · 182 lines

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 · 182 lines · 110 tokens per session scan A 7e7ab0f91076

Subscribe to this mod's changes

atomize is a skill published in the GitHub repository nemock/company-brain (5 stars, last pushed 3mo ago), licensed MIT. It adds 110 tokens to every session and 2,295 once invoked, about $0.0006 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

extract-memory

Extract writing style, personality traits, and key insights from a collection of documents (Markdown, PDF, TXT, DOCX) and save them as a Claude Code memory file. Use when the user wants to capture their writing DNA, build a personal style guide, or create a reusable memory profile from their existing content.

anneheartrecord/charles-skill · 71 tokens

agent-recall

Persistent compounding memory for AI agents. 5 default MCP tools: sessionstart, sessionend, remember, recall, check. Full surface (18 tools) available with --full flag. Two-verb model: inhale (sessionstart) and exhale (sessionend). Correction-first memory with decision trail tracking, watchfor warnings, palace rooms…

Goldentrii/AgentRecall-X · 157 tokens

atlas-absorb

글에서 개념 뽑기. 회의록·기획서·PR 설명에서 개념 후보를 뽑되 중복부터 먼저 확인한다. 사람이 고르기 전에는 쓰지 않는다. "이 글에서 뽑아줘" · "이 문서 반영해줘" 일 때 쓴다.

wlsdks/ontology-atlas · 76 tokens

swarmvault

Use SwarmVault when the user needs a local-first knowledge vault that writes durable markdown, graph, search, dashboard, review, chat-session, context-pack, task-ledger, static AI export, retrieval, and MCP artifacts to disk from books, notes, transcripts, exports, datasets, slide decks, files, URLs, code, and…

swarmclawai/swarmvault · 76 tokens

save

File the current conversation's valuable content (analysis, decision, session summary) as a structured wiki note so it isn't lost in chat history. Triggers on: "save this", "guarda esto", "guárdalo en el wiki", "/save", "$save", "file this", "apunta esto", "guarda esta conversación", "save this session".

sturlese/hippocampus · 78 tokens

kb-ingest

Ingest documents from the workhub vault inbox into the knowledge base. Classifies notes into projects/knowledge/archive, proposes tasks for actionable items, links, and indexes. Use when new notes land in inbox/ or a path needs filing.

atman-33/workhub · 52 tokens