bunkervm CLAUDE.md

Project instructions for BunkerVM, a Python library that runs agent code in disposable, hardware-isolated Firecracker microVMs. They focus on record, replay, snapshots, restoring states, comparing differences, testing, and continuous integration.

In plain words
What is it for?
Guiding code changes, tests, and CI work in the BunkerVM repository, including its two operating modes and record/restore/diff features.
Why use it?
They explain the repository’s current scope and conventions so coding agents can work on its time-travel debugging and MCP support without reintroducing removed components.

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/ashishgituser/bunkervm/claude-md
Clone the repo
git clone --depth 1 https://github.com/ashishgituser/bunkervm
Per session 1,578 This file is loaded in full into every session.
When invoked 1,578 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 $0.01578 $0.01578
Opus 5 $0.00789 $0.00789
Sonnet 5 $0.00316 $0.00316
Haiku 4.5 $0.00158 $0.00158

Measured yesterday against content hash 9b192fd9cd2b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

bunkervm 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 yesterday.

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 · 45 lines

How it starts

The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.

BunkerVM

Hardware-isolated Linux sandbox for AI agents — Firecracker microVMs with record/replay/snapshot/diff. Pip package bunkervm, console script bunkervm, current version 0.9.4. Core dep is just mcp>=1.0.0.

Repo scope (as of the 2026-08 trim-down, see CHANGELOG.md): trimmed to the core differentiated pitch — time-travel debugging (record/snapshot/restore/diff) — plus MCP support. The desktop GUI app and the dedicated LangChain/OpenAI-Agents-SDK/CrewAI integration modules were untested, high-maintenance, and diluted focus; they were moved to backup/ (gitignored, not tracked, still in git history). Don't recreate them without deliberately deciding to re-scope — check backup/README.md first if something seems to be "missing." SecureAgentRuntime.as_tool() / .as_openai_tool() in agent_runtime.py remain as minimal single-tool adapters (lazy-import the upstream SDK directly, no bundled toolkit class).

Two operating modes (pervasive throughout the codebase)

  • Direct mode — talks straight to Firecracker via KVM (Linux only). Sandbox._vm = bunkervm/vm_manager.py::VMManager, comms over vsock UDS via bunkervm/sandbox_client.py. Supports true Firecracker snapshots (instant, full VM state: memory/fs/processes).
  • Engine mode — a background daemon (bunkervm/engine/daemon.py, REST API on localhost:9551) owns the only Firecracker access; everything else (CLI, SDK, desktop, MCP server) is a thin HTTP client via bunkervm/engine_client.py / bunkervm/engine/client.py. Used transparently on Windows via a WSL2 bridge (bunkervm/engine/wsl_bridge.py). README lists only Linux (native /dev/kvm) and Windows+WSL2 as supported; macOS is not a documented supported platform (Firecracker requires KVM, which macOS lacks). Cannot do real VM snapshots — some direct-mode-only features raise RuntimeError here (manual checkpoint(), true snapshot restore()).
  • Sandbox.start() tries _resolve_engine() first, then falls back to _start_direct(). Capability parity between the two modes is incomplete by design, not a bug.

Read the full file on GitHub · 45 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. yesterday First seen · 45 lines · 1,578 tokens per session scan A 9b192fd9cd2b

Subscribe to this mod's changes

bunkervm CLAUDE.md is an instructions file published in the GitHub repository ashishgituser/bunkervm (0 stars, last pushed 14d ago), licensed MIT. It adds 1,578 tokens to every session, about $0.0079 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.

Related

Other instructions, from other repositories