understanding-work

understanding-work is a skill for Claude Code, Codex from emergent-wisdom/understanding-graph. It costs 105 tokens per session (2,845 once invoked), scanned A, original, MIT.

A set of instructions for doing substantial work through an Understanding Graph, a structured place where reasoning and results are kept for later use.

In plain words
What is it for?
It helps with research, writing, coding, design, diagnosis, and decisions when the work should build on a shared graph of understanding.
Why use it?
It prevents important decisions, explanations, and draft material from existing only in a chat that may be forgotten.

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 understanding-graph plugin — 10 skills, 2 MCP servers 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/emergent-wisdom/understanding-graph/understanding-work
Any agent
npx skills add emergent-wisdom/understanding-graph --skill understanding-work
Clone the repo
git clone --depth 1 https://github.com/emergent-wisdom/understanding-graph

Made for: Claude Code, Codex.

Or install understanding-graph, the plugin that ships this one along with the rest of its 10 skills, 2 MCP servers.

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 understanding-work

README.md
[![agentmods](https://agentmods.dev/badge/skills/emergent-wisdom/understanding-graph/understanding-work.svg)](https://agentmods.dev/skills/emergent-wisdom/understanding-graph/understanding-work)
Your own site
<a href="https://agentmods.dev/skills/emergent-wisdom/understanding-graph/understanding-work"><img src="https://agentmods.dev/badge/skills/emergent-wisdom/understanding-graph/understanding-work.svg" alt="Measured on agentmods" height="20"></a>
Per session 105 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,845 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 $0.00105 $0.02845
Opus 5 $0.00053 $0.01422
Sonnet 5 $0.00021 $0.00569
Haiku 4.5 $0.00011 $0.00284

Measured 5d ago against content hash 4769013c8580, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

understanding-work 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 5d 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/understanding-work/SKILL.md · 238 lines

How it starts

The opening of the file, as written. The whole thing — 238 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Work through Understanding

Protocol compatibility: fluid-understanding-v1.

When this skill is active, the graph is the canonical medium in which the task develops. It is not a memory aid or filing cabinet visited after the answer is finished. The primary task is not merely to write a novel, answer a question, or produce code. It is to do that work through the Understanding Graph so the understanding that makes the artifact possible remains available for new work.

Do not wait for the user to name a tool. Begin in the graph unless the user explicitly asks not to use it. Chat may report progress, ask for a decision, or present a result; it does not replace graph work.

Maintain medium integrity: never leave newly developed substantive artifact text or communicable understanding only in chat. Before completing a turn, write the exact artifact units and genuine testimony to the graph. The final response may mirror or summarize what was committed. This is an invariant of the medium, not a mandatory sequence for how understanding must unfold.

Enter the medium

Call project_list before substantive graph work and confirm the active project matches the task. If none is active, select an existing match or create a descriptive, task-scoped one with project_switch; an empty list means create one. If several existing projects are plausible, ask rather than guess. Never create a generic catch-all project.

Choose the amount of navigation aid that suits the model and task:

  • Direct: call graph_understand, graph_batch, or another appropriate graph tool immediately and choose for yourself when to deepen, connect, test, revisit, disrupt, or pause.
  • Guided: call graph_suggest_next when graph-specific pointers could deepen or diversify understanding, recover neglected material, test the current view, or make a useful connection visible.

The live server may provide guided prompts automatically or run in direct mode. Either way, graph_suggest_next remains optional and callable on demand. Its weighted routes are provocations, not quality guarantees or required phases; choose, combine, modify, reject, replace, or skip them. A route may call graph_understand to bring task-relevant graph material into view. Treat any returned material as provisional prior understanding, not authority. If the graph is empty, begin honestly without inventing a past.

Read the full file on GitHub · 238 lines

Files

What ships with it

1 file 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. 5d ago First seen · 238 lines · 105 tokens per session scan A 4769013c8580

Subscribe to this mod's changes

understanding-work is a skill published in the GitHub repository emergent-wisdom/understanding-graph (1 stars, last pushed 11d ago), licensed MIT. It adds 105 tokens to every session and 2,845 once invoked, about $0.0005 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

braindb

Memory recall and save. Use at every conversation start and whenever the user shares personal information, expertise, project context, preferences, or decisions worth remembering long-term.

dimknaf/braindb · 36 tokens

braindb-agent

Persistent memory across sessions via the BrainDB agent. Use at conversation start and whenever you need to recall what you know about the user or save new information to long-term memory.

dimknaf/braindb · 40 tokens

braindb-custom-profile

How to author a BrainDB custom profile — prompt add/replace fragments and an optional keyless ingestor — that shapes wiki naming/structure and feeds a custom ingestion source, with zero effect on defaults when inactive.

dimknaf/braindb · 49 tokens

dev-browser

Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include "go to [url]", "click on", "fill out the form", "take a screenshot", "scrape", "automate", "test the website"…

MemTensor/MemOS · 84 tokens

harness-creator

Build, audit, and improve harnesses that make AI coding agents reliable: AGENTS.md/CLAUDE.md instruction files, feature/state tracking, verification gates, scope boundaries, session handoff, memory persistence, context budgets, tool-permission safety, and multi-agent coordination. Use this whenever a coding agent is…

walkinglabs/learn-harness-engineering · 142 tokens

browserwing-admin

Manage and operate BrowserWing — an intelligent browser automation platform. Install dependencies, configure LLM, create/manage/execute automation scripts, use AI-driven exploration to generate scripts, browse the script marketplace, and troubleshoot issues.

MemTensor/MemOS · 47 tokens