distill

distill is a command for coding agents from extremeclarity/claude-plugins. It costs 7 tokens per session (867 once invoked), scanned A, original, MIT.

A command that reviews conversation knowledge and proposes durable updates to an agent's shared context files.

In plain words
What is it for?
It is for identifying stable project knowledge, checking whether it deserves long-term storage, and presenting proposed context changes for approval.
Why use it?
It helps preserve information that will matter in future sessions while filtering out temporary details and implementation-specific advice.

Command

Installs and runs on its own, but its text points at files inside its plugin — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed. Installing the plugin gets both.

Part of the worldview plugin — 5 commands 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 commands/extremeclarity/claude-plugins/distill
Clone the repo
git clone --depth 1 https://github.com/extremeclarity/claude-plugins

Or install worldview, the plugin that ships this one along with the rest of its 5 commands.

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 distill

README.md
[![agentmods](https://agentmods.dev/badge/commands/extremeclarity/claude-plugins/distill.svg)](https://agentmods.dev/commands/extremeclarity/claude-plugins/distill)
Your own site
<a href="https://agentmods.dev/commands/extremeclarity/claude-plugins/distill"><img src="https://agentmods.dev/badge/commands/extremeclarity/claude-plugins/distill.svg" alt="Measured on agentmods" height="20"></a>
Per session 7 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 867 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 $0.00007 $0.00867
Opus 5 $0.00003 $0.00434
Sonnet 5 $0.00001 $0.00173
Haiku 4.5 $0.00001 $0.00087

Measured 4d ago against content hash e7e07828c90c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

distill 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.

plugins/worldview/commands/distill.md · 101 lines

How it starts

The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Distill

Promote durable knowledge from conversation to worldview.

Execution Model

Silent execution: Do not narrate your reasoning process. Work through steps 1-3 internally, then output only the final plan in step 4.

  1. Enter plan mode
  2. Read — Load ${CLAUDE_PLUGIN_ROOT}/scripts/context-schema.json and all context files. Also read WORLDVIEW.md if it exists.
  3. Plan — Generate proposed changes using distill rules below (with any WORLDVIEW.md overrides applied). Validate against criteria.
  4. Present — Show only the final plan to user for approval (no preamble, no reasoning)
  5. Execute — Apply approved changes

The Bar

When in doubt, don't promote.

Every addition costs context space in every future session. Before promoting anything, ask:

  1. Future Value Test: Will this meaningfully help Claude in future sessions? If you can't articulate a concrete scenario where this knowledge changes Claude's behavior for the better, don't add it.

  2. Conceptual Test: Would this still matter if we completely changed the tech stack? Core knowledge is about WHAT to build and WHY — not implementation details, tool behaviors, or coding patterns.

  3. Uniqueness Test: Is this already captured elsewhere, or is it general knowledge Claude already has? Don't duplicate.


Project Overrides (WORLDVIEW.md)

If WORLDVIEW.md exists, it contains project-specific guidance that overrides or refines the default criteria below. Read it first and apply its instructions when evaluating promotions.

Common overrides:

  • Redefining what a category means for this project
  • Stricter or looser promotion bars
  • Focus areas to prioritize or ignore

Promotion Criteria

→ Terminology

  • Conceptual: About problem space, users, product — NOT technical/implementation terms
  • Recurrent: Term appears multiple times across conversations
  • Relevant: High relevance to project direction
  • Non-obvious: Meaning differs from common usage or is project-specific

Read the full file on GitHub · 101 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 · 101 lines · 7 tokens per session scan A e7e07828c90c

Subscribe to this mod's changes

distill is a command published in the GitHub repository extremeclarity/claude-plugins (4 stars, last pushed 6mo ago), licensed MIT. It adds 7 tokens to every session and 867 once invoked, about $0.0000 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.