fluidbox: Instructions file for Claude Code

CLAUDE.md

fluidbox CLAUDE.md is an instructions file for Claude Code from hrishikeshdkakkad/fluidbox. It costs 15,201 tokens per session, scanned C, original, MIT.

Instructions for Fluidbox, a system that runs AI coding agents in temporary, controlled sandboxes. It covers the system design, commands, cloud deployment, and how each run produces an approved code change and cost report.

In plain words
What is it for?
Installing and checking the project, running demonstrations, deploying it on Google Cloud, and understanding how agent definitions, runs, approvals, and results work.
Why use it?
They help developers preserve the project’s design rules while setting it up, changing its architecture, or diagnosing deployment problems.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions CLAUDE.md; positional $N argument; mentions Claude Code.

This is hrishikeshdkakkad/fluidbox's own configuration. It tells Claude Code how to work on fluidbox itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything fluidbox configures →

Reuse

Borrowing it

Nothing to install: this file belongs to hrishikeshdkakkad/fluidbox. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/hrishikeshdkakkad/fluidbox/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/hrishikeshdkakkad/fluidbox

Made for: Claude Code.

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 fluidbox CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/hrishikeshdkakkad/fluidbox/claude-md.svg)](https://agentmods.dev/instructions/hrishikeshdkakkad/fluidbox/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/hrishikeshdkakkad/fluidbox/claude-md"><img src="https://agentmods.dev/badge/instructions/hrishikeshdkakkad/fluidbox/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 15,201 This file is loaded in full into every session.
When invoked 15,201 The same file — it is already loaded in full.
Security scan C 2 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.15201 $0.15201
Opus 5 $0.07600 $0.07600
Sonnet 5 $0.03040 $0.03040
Haiku 4.5 $0.01520 $0.01520

Measured 8d ago against content hash 003bef3d5ee4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade C, and why

fluidbox CLAUDE.md scanned grade C with 2 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 8d 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.

Cloud metadata endpointhighServer-side request forgery

One request to 169.254.169.254 can return temporary IAM credentials.

- **ALL control-plane egress rides ONE hardened boundary, and admission happens TWICE** (Phase E, Gap 7; `egress.rs`, `netpolicy.rs` in core). Two hardened clients: `state.egress_http` (broker MCP + discovery + delivery

Makes network callslowCapability

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

- **Connector OAuth is now a ONE-TIME, browser-bound flow with REUSABLE client identities** (Phase D, invariant 20; `oauth.rs`, migrations 0015/0016). `POST /v1/connections/{id}/oauth/start` (and catalog Connect) returns
CLAUDE.md · 161 lines

How it starts

The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

What this is

fluidbox is a control plane that runs AI coding agents in governed, disposable sandboxes. A user registers a versioned agent definition; each run freezes an immutable RunSpec, provisions a fresh sandbox, streams a live event timeline, pauses for human approval (or auto-decides in autonomous mode), and ends with a diff + cost report. PLAN.md is the authoritative design doc and roadmap — read it before making architectural changes. It defines convergence invariants (§2) that every change must preserve.

Hard constraints (from PLAN.md, non-negotiable): fluidbox-authored backend is 100% Rust; the Next.js dashboard is presentation-only (all logic in the Rust API); database is Neon Postgres; OSS/MIT.

Commands

just setup          # fresh-clone bootstrap (idempotent): .env + secrets, apps/web/.env.local, pnpm install, runner image
just doctor         # preflight — validates every env gotcha below and prints the exact fix
just demo           # five-minute no-key first-run: deterministic replay through the real gate (isolated: ports 19790/19791/15434, own volume, .demo/); `just demo-down` removes everything
just replay-build   # rebuild the replay runner image after editing images/replay-runner/
just dev            # LiteLLM gateway + server + dashboard together
just server         # Rust control plane only (migrations auto-run on boot)
just web            # dashboard only (Next.js, port 3000)
just gateway-up     # start the pinned LiteLLM container (reads .env)
just sandbox-build  # rebuild the sandbox runner image after editing images/sandbox-runner/
just check          # fmt + clippy -D warnings + test + web build (the full bar)
just e2e            # full acceptance: live demo A + governance + git workspaces + api triggers + schedules + github fan-out + capability catalog + connector catalog/oauth + failure paths (owns the stack; stop `just dev` first)

cargo test -p fluidbox-core                              # fast, no DB needed
cargo test -p fluidbox-db                                # needs DATABASE_URL (local container Postgres)
cargo test -p fluidbox-core policy::tests::               # a single module's tests
cargo run -p fluidbox-cli -- run --task "…" --repo /path # drive a run from the CLI

Read the full file on GitHub · 161 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. 8d ago First seen · 161 lines · 15,201 tokens per session scan C 003bef3d5ee4

Subscribe to this mod's changes

fluidbox CLAUDE.md is an instructions file published in the GitHub repository hrishikeshdkakkad/fluidbox (11 stars, last pushed yesterday), licensed MIT. It adds 15,201 tokens to every session, about $0.0760 per session on Opus 5. A static security scan graded it C with 2 findings (cloud metadata endpoint, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

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.

vercel/next.js · 7,296 tokens

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.

openai/codex · 5,153 tokens

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).

microsoft/vscode · 6,785 tokens

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).

microsoft/vscode · 5,001 tokens

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.

langchain-ai/langchain · 4,469 tokens

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.

github/spec-kit · 7,104 tokens