Borrowing it
Nothing to install: this file belongs to kdpa-llc/mcp-compression-proxy. 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/kdpa-llc/mcp-compression-proxy/main/CLAUDE.mdgit clone --depth 1 https://github.com/kdpa-llc/mcp-compression-proxyWrote 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/kdpa-llc/mcp-compression-proxy/claude-md)<a href="https://agentmods.dev/instructions/kdpa-llc/mcp-compression-proxy/claude-md"><img src="https://agentmods.dev/badge/instructions/kdpa-llc/mcp-compression-proxy/claude-md.svg" alt="Measured on agentmods" 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 | $0.01109 | $0.01109 |
| Opus 5 | $0.00554 | $0.00554 |
| Sonnet 5 | $0.00222 | $0.00222 |
| Haiku 4.5 | $0.00111 | $0.00111 |
Grade A, and why
mcp-compression-proxy 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 6d 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 — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Guidance for Claude Code and other coding agents working in this repository.
What this is
An MCP server that aggregates tools from several backend MCP servers behind one connection, and compresses their descriptions to cut context use.
Two binaries ship:
mcp-compression-proxy(src/index.ts) — the MCP server itselfmcp-cli(src/cli/) — progressive tool discovery backed by a daemon that holds warm connections to the backends
Backend connections live in src/mcp/client-manager.ts; compression state in
src/services/; config loading and ${VAR} expansion in src/config/.
Commands
npm run build # tsc
npm test # full suite (builds first via pretest)
npm run test:unit # unit only
npm run test:coverage # with coverage
npm run lint # eslint, must be 0 problems
npm run typecheck # tsc --noEmit
npm run sync-version # align src/version.ts with package.json
pretest builds before testing on purpose: three integration suites spawn
dist/index.js and dist/cli/index.js as real subprocesses, so a stale or
missing build fails them with MODULE_NOT_FOUND.
Testing
| Location | Covers |
|---|---|
tests/unit/ |
modules in isolation, 18 suites |
tests/integration/ |
spawns the built binaries over stdio |
tests/e2e/ |
full workflows against mocked clients |
tests/e2e-real/ |
real Ollama, excluded from the default run |
src/index.ts, src/cli/index.ts and src/cli/daemon.ts are excluded from
coverage — they need a real process, and are covered by the subprocess
suites instead.
Rules learned from real failures here
Do not assert an LLM's exact wording. tests/e2e-real talks to a 1B
model; asserting its phrasing contained specific keywords blocked a
docs-only PR. Assert the round-trip produced a substantive answer, and log
the rest.
Never leave a timer or listener holding the event loop. Housekeeping
should never be the reason a process cannot exit — unref() intervals, clear
timeouts in a finally, and remove signal listeners on close. The suite runs
with no open-handle warning and no --forceExit; keep it that way, because
--forceExit would only hide a real production leak.
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.
- 6d ago First seen · 103 lines · 1,109 tokens per session scan A 6326bf0fe54b
mcp-compression-proxy CLAUDE.md is an instructions file published in the GitHub repository kdpa-llc/mcp-compression-proxy (5 stars, last pushed yesterday), licensed MIT. It adds 1,109 tokens to every session, about $0.0055 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.
Other instructions, from other repositories
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.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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).
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
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.
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.