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.
npx agentmods add instructions/marlburrow/hivekeep/claude-mdgit clone --depth 1 https://github.com/MarlBurroW/hivekeepWrote 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/marlburrow/hivekeep/claude-md)<a href="https://agentmods.dev/instructions/marlburrow/hivekeep/claude-md"><img src="https://agentmods.dev/badge/instructions/marlburrow/hivekeep/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 | $0.04507 | $0.04507 |
| Opus 5 | $0.02253 | $0.02253 |
| Sonnet 5 | $0.00901 | $0.00901 |
| Haiku 4.5 | $0.00451 | $0.00451 |
Grade A, and why
hivekeep 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.
How it starts
The opening of the file, as written. The whole thing — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hivekeep
Self-hosted platform of specialized AI agents (Agents) for individuals and small groups. Each Agent has a persistent identity, expertise, memory, and tools. Agents share a single continuous session (no "new conversation"), collaborate with each other, spawn sub-Agents for tasks, and execute scheduled jobs.
Documentation map
Read these files before starting any phase. They are the source of truth.
| File | Content |
|---|---|
schema.md |
Complete SQLite database schema (all tables, indexes, virtual tables) |
api.md |
REST API contracts (request/response for every route) + SSE events |
sse.md |
Real-time/SSE cheat sheet — emit↔handle rules, the 8 recurring sync-bug traps, optimistic reconciliation, review checklist. Read before touching SSE or shared state. |
config.md |
All configurable values with env vars and defaults |
structure.md |
Project file tree, naming conventions, imports, i18n, error format |
prompt-system.md |
How the Agent system prompt is assembled (blocks 1-12) |
compacting.md |
Compacting algorithm + memory extraction pipeline |
memory.md |
Memory v2 design — always-injected per-Agent profile document + episodic archive searched via recall; replaces per-message retrieval injection and the scoring heuristics |
queenie.md |
Conversational onboarding spec — the Queenie configurator Agent, vault-centralized secrets, secure-input tools, avatar-style customization (Phase 27) |
files.md |
Files section spec — workspace file browser/editor (tree + tabs + CodeMirror), workspace REST API + workspace:changed SSE, share-to-file-storage, chat integrations (@ file palette, clickable paths) |
interactive-setup.md |
Interactive setup cards spec — generic, declarative in-chat OAuth sign-in + QR pairing (setup methods secret/oauth/qr), extending the secret_prompts card machinery without a DB migration |
external-api.md |
External API spec — machine-to-machine conversational access: declared external clients + API keys (bearer), /api/v1/* send (wait/poll) with requestId correlation, hybrid target (main timeline or full-power isolated thread). Reuses the queue, session-lane and turn-completion machinery. |
testing-instance.md |
Local test instance — a ready, LLM-capable seeded DB (~/.local/share/hivekeep-testdata/, Claude Max provider via host creds) for booting the real app to verify changes. Read it before running the app locally: this shell inherits PROD env (port 3000 + live DB) — scripts/seed-test-db.ts regenerates the seed. |
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.
- 4d ago First seen · 214 lines · 4,507 tokens per session scan A f19ab816a2dd
hivekeep CLAUDE.md is an instructions file published in the GitHub repository MarlBurroW/hivekeep (51 stars, last pushed today), licensed MIT. It adds 4,507 tokens to every session, about $0.0225 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.
Other instructions, from other repositories
LobsterAI AGENTS.md
AGENTS.md instructions for netease-youdao/LobsterAI, covering agents.md, instruction scope, project snapshot, cowork vs openclaw and commands.
orbit CLAUDE.md
Claude Code instructions for schmitech/orbit, covering claude.md, project overview, commands, setup and running the server.
hope-agent AGENTS.md
AGENTS.md instructions for shiwenwen/hope-agent, covering hope agent, 安全红线, 提交前检查（强制）, 分支与发布 and 设置约定.
opencode-chat-bridge AGENTS.md
Instructions for ominiverdi/opencode-chat-bridge: You are a helpful assistant in a chat interface.
golemcore-bot CLAUDE.md
Instructions for alexk-dev/golemcore-bot, covering claude.md — golemcore bot, build & test, modules, architecture and package structure.
telegram-claude-codex CLAUDE.md
Instructions for Mark-Life/telegram-claude-codex, covering claude.md, commands, architecture, module overview (src/) and data flow.