Borrowing it
Nothing to install: this file belongs to munhq/distil. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/munhq/distil/main/CLAUDE.mdgit clone --depth 1 https://github.com/munhq/distilWrote 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.
[](https://agentmods.dev/instructions/munhq/distil/claude-md)<a href="https://agentmods.dev/instructions/munhq/distil/claude-md"><img src="https://agentmods.dev/badge/instructions/munhq/distil/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/munhq/distil/claude-md"><img src="https://agentmods.dev/badge/instructions/munhq/distil/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.01447 | $0.01447 |
| Opus 5.5 | $0.00579 | $0.00579 |
| Sonnet 5.5 | $0.00289 | $0.00289 |
| Haiku 4.5 | $0.00145 | $0.00145 |
Grade A, and why
distil CLAUDE.md 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 2mo 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.
How it starts
The opening of the file, as written. The whole thing — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
distil — context optimization for LLM agents
What this is
A Rust crate and a set of binaries that measure where an agent conversation's tokens go, and compress that conversation only where compression pays for the prompt cache it invalidates.
Read README.md before changing anything here. It carries the measurement that
shapes every design decision: a history rewrite invalidates the cached prefix
from the edit onwards, converting cache reads at 0.1x into cache writes at 1.25x
or 2.0x, and on real traffic that costs more than the rewrite saves. Do not add
a layer or a default that rewrites history without pricing it against that table.
Consumption modes
- Crate — a path or git dependency. Pure library, no binary. Not on crates.io yet; the name is unclaimed.
- MCP server —
distil-mcp, shipped as@munhq/distilon npm, as a Claude Code plugin, and as release binaries for six targets. - HTTP server —
distil-proxy, direct (POST /v1/optimize) or as a proxy in front of an upstream LLM API.
Architecture
Ctx (messages + tools + turn)
-> Pipeline [ Layer, Layer, ... ]
-> PipelineResult (tokens_before, tokens_after, per-layer detail)
Every layer implements Layer from src/pipeline.rs:
apply(&self, ctx: &mut Ctx, counter: &dyn TokenCounter) -> LayerResulthandle_tool_call(name, args) -> Option<String>for layers that inject tools
Layers (src/layers/)
| Layer | File | What it does | Touches history |
|---|---|---|---|
| CacheAlignLayer | cache_layer.rs | Orders content so the stable prefix stays cacheable | no |
| ScratchpadLayer | scratchpad_layer.rs | Keeps working state outside the context window | no |
| RegistryLayer | registry_layer.rs | Compact tool catalog plus a tool_search meta-tool |
no |
| CodeModeLayer | code_mode_layer.rs | JS sandbox that chains several tools in one call | no |
| BudgetLayer | budget_layer.rs | Trims oldest messages to fit a token budget | yes |
| MaskingLayer | masking_layer.rs | Replaces old tool results with summaries, truncates JSON | yes |
| SummarizationLayer | summarization_layer.rs | LLM-based semantic compression | yes |
| CompactionLayer | compactor_layer.rs | Dedup, whitespace strip, merges adjacent messages | yes |
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.
- 2mo ago First seen · 123 lines · 1,447 tokens per session scan A 82e93f21f60d
distil CLAUDE.md is an instructions file published in the GitHub repository munhq/distil (1 stars, last pushed 22d ago), licensed MIT. It adds 1,447 tokens to every session, about $0.0058 per session on Opus 5.5. 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.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.