manifest CLAUDE.md

manifest CLAUDE.md is an instructions file for coding agents from mnfst/manifest. It costs 20,157 tokens per session, scanned A, original, MIT.

Development instructions for Manifest, a service that routes AI-agent requests to language-model providers and tracks requests, costs, and token use.

In plain words
What is it for?
Use it when developing Manifest, preparing a clean PostgreSQL database for a run, or working with its model-routing dashboard and supported AI agents.
Why use it?
It gives developers consistent guidance for running and testing the project, including its cloud setup and database requirements.

Instructions file

About the project

Manifest is an open-source gateway that gives AI agents and applications one OpenAI-compatible endpoint for API keys, subscriptions, local models, and other providers. It routes requests, records query data, tracks usage, and can fall back when a model request fails. The catalogue entry includes instructions, a hook, and a setting for operating Manifest.

mnfst/manifest · 7,506 stars · on GitHub

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/mnfst/manifest/claude-md
Clone the repo
git clone --depth 1 https://github.com/mnfst/manifest

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 manifest CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/mnfst/manifest/claude-md.svg)](https://agentmods.dev/instructions/mnfst/manifest/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/mnfst/manifest/claude-md"><img src="https://agentmods.dev/badge/instructions/mnfst/manifest/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 20,157 This file is loaded in full into every session.
When invoked 20,157 The same file — it is already loaded in full.
Security scan A 1 finding. 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 $0.20157 $0.20157
Opus 5 $0.10079 $0.10079
Sonnet 5 $0.04031 $0.04031
Haiku 4.5 $0.02016 $0.02016

Measured today against content hash c66667bc0923, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

manifest CLAUDE.md scanned grade A with 1 finding 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 today.

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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

→ ProviderModelFetcherService.fetch(providerId, apiKey)
CLAUDE.md · 859 lines

How it starts

The opening of the file, as written. The whole thing — 859 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Manifest Development Guidelines

Last updated: 2026-09-03

What Manifest Is

Manifest is a smart model router for AI agents. It sits between an agent and its LLM providers, scores each request, and routes it to the cheapest model that can handle it. The dashboard tracks logical requests and their provider attempts, costs, and tokens across any agent that speaks OpenAI-compatible HTTP.

"Harness" is the dashboard word for an agent. The UI now labels agents Harnesses (nav item "Harnesses", routes under /harnesses, categories AI agent / App AI SDK / Coding Assistant from CATEGORY_LABELS in packages/shared/src/agent-type.ts). This is a copy-level rename only: backend code, database tables (agents, agent_messages, …), API routes (/api/v1/agents/*), and entity/service names all still say agent. Legacy /agents/* dashboard URLs redirect to /harnesses/*. When writing UI copy say "harness"; when writing code or API docs keep "agent".

Pivot note: Manifest is pivoting toward "the self-healing layer for APIs" (see the README banner). The dashboard shows a sidebar PivotAnnouncement card with a waiting-list modal in every deployment mode (per-session dismiss); waiting-list claims land on POST /api/v1/waitlist/pivot/claim and record their origin (cloud vs self-hosted). The open-source gateway remains available and maintained.

Supported agents: see AGENT_PLATFORMS in packages/shared/src/agent-type.ts for the current list (OpenClaw, Hermes, Claude Code, OpenCode, generic OpenAI/Anthropic SDK slots, and others — don't duplicate the list here, it grows independently of this doc). OpenClaw remains the deepest integration, but no new code or copy should frame Manifest as OpenClaw-only. When adding examples, prefer "AI agent" as the noun and pick OpenClaw as the worked example rather than the sole target. Manifest is consumed as a generic OpenAI-compatible HTTP endpoint — there are no first-party OpenClaw plugins in this repo anymore.

Read the full file on GitHub · 859 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. today Changed · +5 lines · +317 tokens per session c66667bc0923
  2. yesterday Changed · +40 lines · +1,733 tokens per session fa81b9395d86
  3. 2d ago Changed · +3 lines · +103 tokens per session dd3a204e4af2
  4. 5d ago First seen · 811 lines · 18,004 tokens per session scan A 41ef9a33e899

Subscribe to this mod's changes

manifest CLAUDE.md is an instructions file published in the GitHub repository mnfst/manifest (7,506 stars, last pushed today), licensed MIT. It adds 20,157 tokens to every session, about $0.1008 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.