openchief CLAUDE.md

openchief CLAUDE.md is an instructions file for coding agents from serpin-taxt/openchief. It costs 6,349 tokens per session, scanned A, original, MIT.

A project instruction file for OpenChief, a serverless system whose AI agents watch business tools and create reports. It describes the monorepo, Cloudflare services, data connectors, dashboard, database, and AI components.

In plain words
What is it for?
For working on business-data connectors, event processing, agent memory, report generation, the React dashboard, and Cloudflare deployment.
Why use it?
It gives agents the system map needed to change one service without overlooking related queues, storage, connectors, or reports.

Instructions file

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 instructions/serpin-taxt/openchief/claude-md
Clone the repo
git clone --depth 1 https://github.com/serpin-taxt/openchief

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 openchief CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/serpin-taxt/openchief/claude-md.svg)](https://agentmods.dev/instructions/serpin-taxt/openchief/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/serpin-taxt/openchief/claude-md"><img src="https://agentmods.dev/badge/instructions/serpin-taxt/openchief/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 6,349 This file is loaded in full into every session.
When invoked 6,349 The same file — it is already loaded in full.
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 $0.06349 $0.06349
Opus 5 $0.03174 $0.03174
Sonnet 5 $0.01270 $0.01270
Haiku 4.5 $0.00635 $0.00635

Measured 4d ago against content hash ebb9b635e393, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

openchief 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 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.

CLAUDE.md · 454 lines

How it starts

The opening of the file, as written. The whole thing — 454 lines — stays where its author put it; the contents beside it link to each section on GitHub.

OpenChief

AI agents that passively watch business tools and produce reports via Claude. Open-source, serverless, runs on Cloudflare Workers.

Architecture

  • pnpm monorepo + Turborepopackages/shared, workers/*, workers/connectors/*
  • Cloudflare Workers — runtime (Durable Objects), router (Queue consumer), dashboard (React SPA + API), connectors (one Worker per data source)
  • Cloudflare D1 — shared database for agents, events, reports, revisions, identity mappings, model settings
  • Cloudflare KV — caching layer for subscriptions, reports, user profiles, connector metadata, avatars
  • Cloudflare Queuesopenchief-events queue connects connectors to the router
  • Cloudflare Vectorize — vector index for RAG (long-term memory), stores report/event embeddings
  • Cloudflare Workers AI — embedding model (@cf/baai/bge-base-en-v1.5) for Vectorize indexing and retrieval
  • Durable Objects — per-agent persistent state with SQLite inbox (events, reports, chat history, reasoning log)
  • React + Tailwind CSS v4 — SPA dashboard at workers/dashboard
  • Anthropic Claude — LLM for report generation and agent chat (claude-sonnet-4-6 default)

Repo Structure

openchief/
├── packages/shared/          # @openchief/shared — types, matching, ULID generation
├── workers/
│   ├── runtime/              # Agent Durable Object runtime (report generation + chat)
│   ├── router/               # Event router (queue consumer → D1 persistence)
│   ├── dashboard/            # React SPA + API worker (all UI + management endpoints)
│   ├── demo-engine/          # Synthetic event generator for demo mode (Serpin's Burger Shack)
│   └── connectors/           # One Worker per data source (14 connectors, all implemented)
│       ├── github/           # Webhook + polling
│       ├── slack/            # Webhook + polling + backfill + identity sync
│       ├── discord/          # Webhook (Ed25519) + polling
│       ├── figma/            # OAuth + webhook + polling (file updates, versions, comments, deletes, library publishes)
│       ├── jira/             # Polling (issues, transitions, sprints)
│       ├── jpd/              # Polling (Jira Product Discovery — ideas, insights)
│       ├── notion/           # Polling (pages, databases, comments)
│       ├── intercom/         # Webhook (HMAC-SHA1) + polling (conversations)
│       ├── twitter/          # OAuth PKCE + polling (multi-account, search)
│       ├── amplitude/        # Polling (metrics snapshots)
│       ├── google-calendar/  # OAuth + polling (calendar events)
│       ├── google-analytics/ # Polling (GA4 metrics via service account)
│       ├── quickbooks/       # OAuth (Intuit) + polling (invoices, payments, P&L)
│       └── rippling/         # Polling (employees, org structure, time-off)
├── agents/                   # 15 agent JSON definitions (data, not code)
├── migrations/               # 7 D1 SQL migration files
├── scripts/                  # setup.ts, seed-agents.ts, generate-config.ts, deploy.ts, teardown.ts
├── openchief.example.config.ts
├── turbo.json
└── pnpm-workspace.yaml

Read the full file on GitHub · 454 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 · 454 lines · 6,349 tokens per session scan A ebb9b635e393

Subscribe to this mod's changes

openchief CLAUDE.md is an instructions file published in the GitHub repository serpin-taxt/openchief (24 stars, last pushed 5mo ago), licensed MIT. It adds 6,349 tokens to every session, about $0.0317 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 2026-08-30.