Show HN: Melra – pause ambiguous MCP writes instead of blindly retrying

XAGI-Lab released MELRA, an open-source local execution runtime that governs AI agent tool calls routed through it by checking each call against policy, requesting approval when needed, verifying declared evidence, and keeping a redacted receipt of what runs to an outcome. MELRA dispatches mutations at most once per task and parks uncertain outcomes for reconciliation instead of replaying them, though the project states exactly-once external effects still depend on the provider and calls made through other tools remain outside MELRA's control. The alpha software requires Node.js 22.13 or newer, or 23.4 or newer, installs via `npx @melra/cli@alpha setup`, and writes itself into the config of a detected MCP client such as Claude Code, Cursor, or Gemini CLI, with APIs subject to change before version 1.0.

M odular E xecution L ayer for R eliable A utonomy Your AI agent decides what to do. MELRA governs how tool calls are authorized, verified, and recovered. An open-source local execution runtime for agent tool calls you route through MELRA. It checks each call against policy, requests approval when needed, verifies declared evidence, and keeps a redacted receipt of what runs to an outcome. For each task, MELRA dispatches mutations at most once; uncertain outcomes are parked for reconciliation instead of replay. Exactly-once external effects still depend on the provider. Calls made through other tools remain outside MELRA's control. 26-second demo 26-second-demo · Quick start quick-start · How it works how-it-works · See what it is doing see-what-it-is-doing · Docs documentation · Contribute contributing · Discussions https://github.com/XAGI-Lab/melra/discussions · Product Hunt launch melra-product-demo.mp4 Warning MELRA is alpha software. The local runtime is tested end to end, but APIs may change before 1.0 . Use an isolated workspace, keep domain and command allowlists narrow, and read every approval before you give it. You need Node.js 22.13 or newer, or 23.4 or newer . From the project the agent should work in, one command sets everything up: cd ~/project npx @melra/cli@alpha setup That directory becomes the agent's workspace, the boundary for its file and command operations. setup shows it, and asks before making your home directory, a directory holding it, or / the workspace; set MELRA WORKSPACE to choose another. In a terminal, setup is a short wizard. It writes a safe local policy, finds the agent apps on your machine, asks which one should use MELRA, and writes MELRA into that app's own config file, keeping everything else in it and saving a .bak first : Which MCP client should use MELRA? 1 Claude Code detected: 'claude' on PATH 2 Cursor detected: config file found 3 Gemini CLI detected: config folder found ... Choose 1-26 1 : 1 Write it into ~/project/.mcp.json? Y/n ✓ Workspace ~/project ✓ Policy ~/.melra/policy.json ✓ Claude Code ~/project/.mcp.json Next: Start claude in this workspace and approve the 'melra' server when it asks; 'claude mcp list' then shows it connected. Then restart your agent app Claude Code asks you to approve a project's server once and ask it to do something, for example "use melra to show me the system info" . That's it. Prefer flags, a container, or building from source? | How | Command | |---|---| | One client, no questions | npx @melra/cli@alpha setup --client cursor --write | | Print the config instead of writing it | npx @melra/cli@alpha setup --client codex | | Install globally | npm install -g @melra/cli@alpha && melra setup | | Container no Node needed | docker run --rm ghcr.io/xagi-lab/melra:alpha doctor | | Prebuilt release | download from releases https://github.com/XAGI-Lab/melra/releases , verify against SHA256SUMS , then node melra/dist/bin.js doctor | | From source | git clone https://github.com/XAGI-Lab/melra.git && cd melra && corepack enable && pnpm install --frozen-lockfile && pnpm build && pnpm melra setup | | One file, no Node needed experimental | from a checkout: pnpm build:sea , then ./dist-sea/melra setup ; see single executable https://github.com/XAGI-Lab/melra/blob/main/docs/INSTALLATION.md single-executable-experimental | | Check what you installed | npm audit signatures verifies registry signatures; npm provenance links each package to its GitHub Actions build see npm packages https://github.com/XAGI-Lab/melra/blob/main/docs/INSTALLATION.md npm | Every command prints readable output in a terminal and plain JSON when piped or with --json , so scripts keep working. MELRA creates <MELRA HOME /payload.key with private permissions on first start; back it up with the SQLite files, because losing it makes persisted payloads unreadable. See installation https://github.com/XAGI-Lab/melra/blob/main/docs/INSTALLATION.md for every install path, environment variable, and the hardened Docker invocation. Caution This project publishes only under the @melra/ npm scope @melra/cli is the CLI and on its releases https://github.com/XAGI-Lab/melra/releases , where the Python SDK's wheel is. An unscoped npm package melra or a PyPI package melra is not this project's . Install from that scope, from this repository, or from the releases. If MELRA helps, star the repository https://github.com/XAGI-Lab/melra so other agent builders can find it. Trying it in Cursor or another client? Share the client and its version, your OS, and your MELRA version npx @melra/cli@alpha version in issue 184 https://github.com/XAGI-Lab/melra/issues/184 ; partial test results are useful too. AI agents no longer just talk. They edit files, run commands, click through websites and call APIs on your behalf. Four questions come with that, and most agents answer them differently, or not at all: | Every action MELRA receives is checked against your policy, and checked again at the moment it runs. Risky changes wait for your exact approval. | "No error" is not success. An action counts only when the evidence you declared holds: the file exists, the exit code is 0, the page says what it should. | Every mutation MELRA dispatches is recorded before it runs and runs at most once . After a crash, MELRA settles it from its evidence or the provider; what it cannot settle waits in recovery required and is never replayed. | Every action MELRA runs to an outcome leaves a redacted, hash-linked receipt , so you can audit what ran, who asked, and why it was allowed. | MELRA answers those questions in one place for tool calls routed through it, across supported agents. It sits underneath your agent for those calls, the way an operating system sits underneath your apps: The LLM reasons. The harness manages the loop. MELRA governs the effect lifecycle for calls routed through it. Swap the model or the agent; MELRA's policies, approvals, credentials, history and receipts can stay in place. MELRA speaks the Model Context Protocol https://modelcontextprotocol.io , so any MCP client can use it. For these, melra setup --client <name writes the config in the client's own format and location, and melra doctor tells you whether each one is pointed at MELRA: | Agent | --client | Config it writes | |---|---|---| | Claude Code | claude-code | <project /.mcp.json | | Claude Desktop | claude-desktop | claude desktop config.json | | OpenAI Codex CLI | codex | ~/.codex/config.toml | | Cursor | cursor | ~/.cursor/mcp.json | | 🟦 VS Code Copilot agent mode | vscode | <project /.vscode/mcp.json | | Gemini CLI | gemini | ~/.gemini/settings.json | | Windsurf | windsurf | ~/.codeium/windsurf/mcp config.json | | Zed | zed | Zed settings.json comments kept | | 🧩 Cline | cline | Cline's cline mcp settings.json | | ⬛ OpenCode | opencode | ~/.config/opencode/opencode.json | | 👻 Kiro | kiro | ~/.kiro/settings/mcp.json | | 🪿 Goose | goose | ~/.config/goose/config.yaml | | Continue | continue | <project /.continue/mcpServers/melra.yaml | | ⚡ Amp | amp | ~/.config/amp/settings.json | | 🦞 OpenClaw | openclaw | ~/.openclaw/openclaw.json mcp.servers | | ⚕️ Hermes Agent Nous Research | hermes | ~/.hermes/config.yaml mcp servers | | GitHub Copilot CLI | copilot | ~/.copilot/mcp-config.json | | 🐉 Qwen Code | qwen | ~/.qwen/settings.json | | Auggie Augment | auggie | ~/.augment/settings.json | | Warp | warp | ~/.warp/.mcp.json | | Junie JetBrains | junie | ~/.junie/mcp/mcp.json | | 🤖 Factory Droid | droid | ~/.factory/mcp.json | | 🌙 Kimi Code | kimi | ~/.kimi-code/mcp.json | | 🪁 Kilo Code | kilo | ~/.config/kilo/kilo.jsonc comments kept | | 🧪 LM Studio | lmstudio | ~/.lmstudio/mcp.json | | 🔌 Anything else | generic | prints the standard mcpServers block | Building your own agent? Use the TypeScript SDK https://github.com/XAGI-Lab/melra/blob/main/packages/sdk-ts or the Python SDK https://github.com/XAGI-Lab/melra/blob/main/sdk-py . Both can spawn MELRA or connect to one that is already running over loopback HTTP or a private Unix socket. For agent frameworks, see the framework examples https://github.com/XAGI-Lab/melra/blob/main/examples/frameworks/README.md : LangChain / LangGraph, the OpenAI Agents SDK, the Vercel AI SDK, and the official MCP SDKs. Note A named client is marked verified only after the released package , not a source checkout, passes discovery, planning, approval, execution, cancellation and receipt retrieval in that client. Per-client status lives in COMPATIBILITY.md https://github.com/XAGI-Lab/melra/blob/main/docs/COMPATIBILITY.md . An effect is one bounded operation an agent routes through MELRA: file.write , terminal.run , browser.click , http.request , and so on. MELRA handles each routed effect in nine stages, split between a plan step and an execute step for a single task, melra plan then melra execute so that nothing runs at planning time : Here is one change from start to finish, in developer mode. In enforced mode https://github.com/XAGI-Lab/melra/blob/main/docs/INSTALLATION.md deployment-modes the challenge carries no phrase: you approve with melra approve <task-id outside the conversation, and the agent's melra execute sends no approval at all. sequenceDiagram autonumber participant A as Your agent participant M as MELRA participant You participant W as Your system A- M: melra plan: write notes.md, evidence "file contains 'done'" M- M: type, classify, check policy, record M-- A: awaiting approval + challenge "APPROVE 3f9c…" A- You: May I write notes.md? You-- A: yes A- M: melra execute with the exact phrase M- M: policy again, not already done? M- W: write the file under a budget M- W: read it back, independently M-- A: verified success + receipt What the final status means , and what an agent should do next: | Status | Meaning | Next step | |---|---|---| | 🟢 verified success | It ran and every declared check passed | Done | | 🟡 partial | It ran, but the evidence did not hold | Look at the evidence, then decide | | 🔴 policy blocked | Refused before anything ran | Read the reason; the same request will be refused again | | 🔵 awaiting approval | A change is waiting for your approval | Developer mode: approve it from the agent that planned it. Enforced mode: melra approve <task-id | | 🟣 recovery required | A change may or may not have happened | Check the world first ; never re-run blindly | | ⚪ failed · cancelled · budget exhausted | It did not complete | Safe to plan again | Show the request an agent actually sends The agent resolves "run the checks" into one bounded operation, plus the evidence that would prove it worked: { "goal": "Run the repository checks", "operation": { "kind": "terminal", "action": "run", "command": "pnpm", "args": "check" }, "requiredEvidence": { "type": "exit code", "value": 0 } } The task reaches verified success only if the process exits 0 . A process that runs and exits 1 completed the action but failed the evidence, so it is reported as partial . MELRA never receives a goal like "fix the production server" : that needs judgement, which belongs to the model. A kernel that interpreted sentences would be a kernel whose guarantees depended on a model. When an agent connects, MELRA also sends it a short guide to this loop the MCP server's instructions , so a model knows to declare evidence, read the status, and never blindly retry an unknown outcome, without anyone prompt-engineering it. You can see each action MELRA handled and its policy decision, and, for each that ran to an outcome, its evidence and receipt, from the terminal or the browser. Nothing here can approve or start work, and approval phrases are never displayed. | From the terminal bash $ melra status MELRA 0.3.0-alpha.14 developer mode Tasks 3 total: 1 awaiting approval, 1 policy blocked, 1 verified success Waiting 1 waiting for approval HTTP 127.0.0.1:7457 not running $ melra tasks ID STATUS OPERATION UPDATED 64199678 awaiting approval file.write 0s ago ca4aec7a policy blocked terminal.run 1s ago a1285e0f verified success system.info 1s ago $ melra watch live, until Ctrl-C 6:07:21 PM task 882cf55c verified success 6:07:22 PM task f813364c awaiting approval melra doctor checks Node, SQLite, the browser and policy, and reports which of your agent apps point at MELRA, with the one command that fixes anything wrong. | In the browser: melra serve --http --open An overview, a searchable task list, and durable workflows with a live event stream. It works at phone width and in light and dark, and it is served with a strict Content-Security-Policy. | | | Default | |---|---| | 🏠 | Local only. stdio, loopback HTTP behind a token, or a mode- 0600 Unix socket only your OS user can open. No account and no hosted service. | | 📴 | Telemetry is off. | | ✍️ | Changes need two things: declared evidence and an exact, expiring, task-scoped approval: a phrase in developer mode, an operator's melra approve in enforced mode. | | 📁 | Paths stay inside your workspace , including terminal working directories; symlink escapes are refused. | | 🚫 | Shells, sudo / su and arbitrary desktop key names are denied. Terminal commands must be on the allowlist by name. | | 🌐 | Private, link-local and cloud-metadata addresses are blocked by the browser and HTTP adapters themselves, after DNS resolution, before any allowlist is consulted; loopback too in enforced mode, unless the policy opens it. | | | Web page content is marked untrusted and can never change policy. | | 🔑 | Secrets are redacted before anything is stored, and agents get capabilities, never your raw credentials. | | ⏱️ | Every call is bounded by a budget and a cancel signal; a stalled browser fails with a clear error instead of hanging. | Unsafe-local mode and the two deployment modes Every row above can be switched off at once with melra serve --unsafe-local or MELRA UNHINGED=1 : no policy, no approvals, no evidence requirement, no workspace confinement, no destination checks. The agent gets exactly the reach your OS user has. The mode announces itself on stderr, in melra doctor and in melra capabilities , and receipts still record what ran. See unsafe-local mode https://github.com/XAGI-Lab/melra/blob/main/docs/INSTALLATION.md unsafe-local-mode . The honest gap in a default install is that MELRA governs the effects it is asked for . An agent holding both a MELRA terminal and a native one makes the kernel optional. MELRA MODE is where you say which situation you are in: | Mode | Meaning | |---|---| | developer default | MELRA governs what it is asked for; the agent may have another path, and you accept that. This is why MELRA can be tried in thirty seconds. | | enforced | You assert the alternative is gone: the agent is sandboxed, holds no privileged secrets, and reaches these systems only through here. | MELRA cannot verify that assertion from inside its own process, and it does not pretend to; the isolation is the operating system's job. What enforced mode does is close every door MELRA owns: --unsafe-local is refused outright, nothing binds outside loopback a 0600 Unix socket is allowed , no client can register itself, and mode is stamped on every receipt. For the operating-system side, examples/deploy https://github.com/XAGI-Lab/melra/blob/main/examples/deploy/README.md has a hardened Docker run and systemd unit, each checked in CI, and melra doctor reports which restrictions are actually in effect. See deployment modes https://github.com/XAGI-Lab/melra/blob/main/docs/INSTALLATION.md deployment-modes , the threat model https://github.com/XAGI-Lab/melra/blob/main/docs/THREAT MODEL.md and the security policy https://github.com/XAGI-Lab/melra/blob/main/SECURITY.md . An LLM decides what to do. A harness runs the loop that gets it there. Something still has to decide whether the resulting effect is allowed, avoid replaying uncertain mutations, prove what happened, and survive a crash in the middle. Today that machinery is rebuilt inside every agent harness, and it dies with the harness. MELRA is that machinery, factored out and made agent-independent. | 🚫 Typical tool server | ✅ MELRA | |---|---| | | | Next to the two layers it is most often mistaken for: | | Agent harness | MCP server | MELRA | |---|---|---|---| | Model loop and reasoning | ✓ | ✗ | ✗ | | Tools | ✓ | ✓ | via adapters | | Agent-independent policy | usually ✗ | possible | core | | Durable effect state | sometimes | rarely | core | | Idempotency across restarts | varies | rare | core | | Crash recovery without replay | varies | rare | core | | Independent verification | uncommon | possible | core | | Evidence receipts | uncommon | possible | core | | Works across agents | ✗ | ✓ | ✓ | | Credential isolation | varies | varies | core | | Hard capability boundary | varies | usually ✗ | partly shipped | Plenty of MCP servers can do these things; this is not a claim that none of them do. The difference is that in MELRA they are the execution contract rather than a per-tool option. The boundary is partly shipped : MELRA MODE=enforced closes every door MELRA owns, but the OS isolation that stops a harness reaching the filesystem around the kernel is the operator's to build; runnable hardened deployment recipes are still to come 52 https://github.com/XAGI-Lab/melra/issues/52 . The feature test. Anything that would still make sense if the request came from ordinary deterministic software belongs in MELRA: authorisation, idempotency, credential isolation, recovery, verification, effect history. Prompt optimisation, LLM memory, model selection, a planner and agent personality belong to the harness above. That is why nothing in MELRA's execution path ever calls a model. The kernel is the contract; these are its shipped implementations. Every one passes through the same policy, approval, budget, verification, receipt and certificate pipeline; an adapter that skipped it would not be an adapter. | | Adapter | What is implemented | |---|---|---| | 🗂️ | Files | Root-confined read, hash, atomic write, move, mkdir and delete, with symlink-escape defenses | | 💻 | Terminal | Shell-free foreground and supervised background processes with allowlists, traits, timeouts, interactive input, cancellation and redaction | | 🌐 | Browser | Isolated Playwright sessions, semantic DOM targets, bounded artifacts, network and popup policy, opt-in profiles, condition-based settling, and every browser call bounded so a stalled page fails clearly | | 🖥️ | Computer | Capability discovery, accessibility-tree and OCR inspection, and governed screenshot, pointer, drag, keyboard and scroll on macOS, Windows and supported Linux/X11 setups | | 📡 | HTTP | GET / HEAD as reads and every other method as an approval-gated at-most-once mutation, pinned to the address the destination check resolved, with credentials injected by the kernel rather than handed to the caller | | 🧠 | Operational memory | A kernel service, not an adapter: scoped SQLite storage of what MELRA knows about its own effects, with hybrid ranking, freshness, expiry, supersession, provenance and redaction | Database, cloud and SaaS adapters are designed and on the roadmap https://github.com/XAGI-Lab/melra/blob/main/ROADMAP.md , not shipped today. Semantic memory about the user preferences, conversation history belongs to the harness, not here. Eleven tools in front of the adapters and one durable workflow controller. The CLI, both SDKs, loopback HTTP and the Unix socket reach the same eleven operations; this list is the kernel contract, not an MCP-specific API. | Tool | Purpose | |---|---| | melra capabilities | Discover operations, platform support, limits and policy posture | | melra plan | Validate, persist and policy-check one bounded operation | | melra execute | Execute an approved plan and verify the declared outcome | | melra task status | Read durable task state | | melra task cancel | Cooperatively cancel pending or running work | | melra receipt | Retrieve redacted evidence and the execution certificate | | melra workflow plan | Validate, preflight, encrypt and persist a bounded workflow | | melra workflow advance | Execute one ready scheduling wave | | melra workflow status | Read the durable workflow projection | | melra workflow cancel | Cooperatively cancel nonterminal workflow work | | melra workflow control | Pause, resume or suspend a run without losing its place | Prefer tool names your model already knows? Set MELRA HARNESS TOOLS=1 and thirteen more appear alongside these read file , write file , run command , browser click , approve , … . They are not a second path: each one builds an ordinary task and runs the same pipeline. See ordinary tool names https://github.com/XAGI-Lab/melra/blob/main/docs/INSTALLATION.md ordinary-tool-names . Workflows compose operation , approval , condition , parallel , bounded loop , checkpoint , compensation , human input and delegation nodes, and every effect still travels through the task pipeline. Each CLI call below is a new process, so this exercises restart persistence with no daemon: melra workflow plan --definition examples/workflows/restart-safe.json save the id melra workflow advance <workflow-id melra workflow advance <workflow-id exits 3 with the write's approval challenge developer mode echoes the phrase; enforced mode runs melra approve <task-id instead melra workflow advance <workflow-id --approval '<approval-id :<exact phrase ' melra workflow advance <workflow-id planned stateVersion=2 running stateVersion=4 inspect=verified complete awaiting approval stateVersion=7 phrase="APPROVE <digest prefix " developer mode running stateVersion=10 write=verified complete verified complete stateVersion=14 checkpoint=verified complete Stop the process after any line and the next command resumes from the same MELRA HOME . Interrupted reads may retry; interrupted changes are never silently repeated. They are settled by re-observation, or they park in recovery required . Everything shipped in this release | Shipped in 0.3.0-alpha.11 and 0.3.0-alpha.12 | Evidence | |---|---| | Restart-safe bounded workflows across nine node kinds | A real MCP process is stopped, replaced and resumed in E2E | | Encrypted exact task, workflow and result payloads | AES-256-GCM storage plus plaintext-leak checks across SQLite/WAL and public projections | | Recovery without silent mutation replay | 8/8 deterministic recovery scenarios, zero duplicates, zero false success | | Loopback HTTP, a 0600 Unix socket, an event stream and a read-only console | Same runtime as stdio, token on every route, 405 on any non- GET to the JSON API | | A client authenticates itself and is named on every receipt | OAuth 2.1 registration through approval to a token | | Two deployment modes | --unsafe-local and a non-loopback bind are refused under MELRA MODE=enforced | | HTTP and API calls as a first-class effect | A mutation whose reply never arrived is reported unknown, not failed | | Agents receive capabilities, not credentials | The kernel holds the secret and scopes it by host and capability; grants carry a size and an expiry | | Evidence says how it is known | execution / state / independent / semantic strength on every evidence item | | Independent-channel verification | http resource matches executes through POST and verifies through a separate GET | | Your own evidence types | A registered verifier answers { "type": "custom" } evidence; planning refuses an unregistered one, and custom evidence is always the weakest strength | | Setup for 26 agent apps | melra setup writes each app's own format; Claude Code, OpenCode, and Hermes Agent connected to a packed release on a real Mac | | OS isolation you can check | melra doctor reports capabilities, no new privs , seccomp, and a read-only root; CI runs it under the hardened Docker and systemd recipes | | A self-authored governance benchmark | GovBench: 362 cases through the real runtime, safety 312/312 , utility 50/50 | See the full CHANGELOG https://github.com/XAGI-Lab/melra/blob/main/CHANGELOG.md . Numbers come from committed scripts and JSON artifacts, with dataset hashes and claim boundaries. They are component measurements , not a claim that MELRA is "the best" or that unlike benchmarks are comparable. | | What | Result | What it means | |---|---|---|---| | ✅ | Safety and execution evals | 52/52 | Deterministic policy, capability-grant, credential-scope, traversal, terminal, memory, computer, HTTP, cancellation and verification scenarios | | 🔁 | Durable Core eval | 8/8 | 100% expected recovery, 0 duplicate execution, 0 false success, 100% event consistency | | 🛡️ | GovBench https://github.com/XAGI-Lab/melra/blob/main/docs/research/GOVBENCH.md self-authored | 312/312 safety, 50/50 utility | 362 model-free cases through the real runtime: containment, approval integrity, verification honesty, exactly-once, redaction, and false positives on legitimate work. Its first run found two real bugs 293/307 , both fixed; governance adds about 0.5–1.4 ms per action | | 🧠 | LoCoMo retrieval | 0.7597 coverage@20 | 1,982 evidence-bearing questions with zero model, embedding or network calls; re-run on 2026-10-14 and identical on every metric | | 🧠 | Synthetic recall | 100/100 Recall@1 | Planted-fact regression over 1,000 records | | 🌐 | Static-page settle | 183.7 ms p50 vs 301.3 ms | 39% less waiting with identical 10/10 correct reads | | 🌐 | Slow-render settle | 10/10 vs 0/10 correct | Condition-based waiting sees the final DOM; a fixed 300 ms wait reads too early | | 💻 | Terminal | 30/30 verified | Shell-free process launch, 48.1 ms p50 on the measured machine | | 🖥️ | Computer control plane | 30/30 probes | 0.032 ms p50 adapter discovery; this is not a desktop task-success score | Raw artifacts: core microbenchmarks https://github.com/XAGI-Lab/melra/blob/main/docs/research/results/core-microbenchmarks.json , LoCoMo https://github.com/XAGI-Lab/melra/blob/main/docs/research/results/locomo-retrieval.json and GovBench https://github.com/XAGI-Lab/melra/blob/main/docs/research/results/govbench-0.3.0-alpha.12-2026-10-14.json . The repository also contains a pre-registered browser-agent harness MiniWoB-125 and the WebArena-Verified Hard-30 registered subset whose task IDs, upstream revisions and dataset hashes are frozen. Its score is not yet claimed : it will be published only after a full fixed-denominator run passes the publication gate. MELRA has not run an official OSWorld, WebArena or LongMemEval submission. Reproduce the scores the full validation gate pnpm check && pnpm evals && pnpm e2e && pnpm pack:check && pnpm security:audit pnpm --filter @melra/evals evaluate:durable-core -- --publishable memory, browser, terminal and computer microbenchmarks pnpm benchmark:core governance benchmark: 362 model-free cases, loopback fixtures only pnpm benchmark:govbench LoCoMo dataset not vendored: CC BY-NC 4.0 git clone https://github.com/snap-research/locomo.git /tmp/locomo pnpm benchmark:locomo -- --dataset /tmp/locomo/data/locomo10.json \ --output docs/research/results/locomo-retrieval.json browser-agent harness contract checks no model, no network pnpm benchmark:browser:check pnpm benchmark:browser:verify-upstream hardened container and a real MCP smoke test docker build -t melra:local . && docker run --rm melra:local doctor && pnpm docker:smoke The MiniWoB suite also needs the pinned MiniWoB++ assets and an agent configuration; see the research index https://github.com/XAGI-Lab/melra/blob/main/docs/research/README.md and the methodology https://github.com/XAGI-Lab/melra/blob/main/docs/research/METHODOLOGY.md . Run outputs HAR, screenshots, video, transcripts are git-ignored and must never be committed. | SDKs Both SDKs spawn a local server or connect to a running one over HTTP or a Unix socket. The token is always passed explicitly, and plain HTTP to a remote host is refused unless you opt in. | CLI at a glance | | | Package | |---|---| | TypeScript https://camo.githubusercontent.com/5d834fe2323dcb6410f02ff43d268073724a7402676a1bbb7829c2eeabedf701/68747470733a2f2f63646e2e73696d706c6569636f6e732e6f72672f74797065736372697074 | @melra/sdk https://github.com/XAGI-Lab/melra/blob/main/packages/sdk-ts | | Python https://camo.githubusercontent.com/a1e513fed558ebbe4d2baecb724cd6e821fa1e91cd706801d62adc06815fcb1a/68747470733a2f2f63646e2e73696d706c6569636f6e732e6f72672f707974686f6e | melra https://github.com/XAGI-Lab/melra/blob/main/sdk-py typed, 3.11–3.14 | | 📜 JSON contracts | @melra/protocol https://github.com/XAGI-Lab/melra/blob/main/packages/protocol , @melra/receipt-schema https://github.com/XAGI-Lab/melra/blob/main/packages/receipt-schema | | Goal | Command | |---|---| | Guided setup | melra setup | | Check this machine | melra doctor | | Serve stdio / HTTP / socket | melra serve · --http --open · --socket <path | | See activity | melra tasks · watch · status | | Run one task | melra run --request task.json | | Approve a task an agent planned | melra approve <task-id · in a browser: melra console | | Inspect a receipt | melra inspect <task-id | | Test a policy | melra policy test --request task.json | | Check conformance | melra conformance | Repository map apps/cli/ CLI and stdio entrypoint kernel services packages/protocol/ strict task, workflow, event and operation schemas packages/runtime-core/ task/workflow lifecycle, events, recovery packages/policy-core/ local policy and scoped approvals packages/storage-sqlite/ transactional tasks, workflows, events, evidence packages/verifier-core/ deterministic evidence predicates packages/receipt-schema/ receipts and execution certificates packages/memory/ operational memory: scoped retrieval and lifecycle packages/server/ runtime composition, MCP, loopback HTTP and Unix socket reference effect adapters packages/file-runtime/ confined filesystem operations packages/terminal-runtime/ shell-free process supervision packages/browser-runtime/ isolated browser automation and stable-DOM wait packages/computer-runtime/ governed local computer-use adapters packages/http-runtime/ governed HTTP and API calls clients and evidence packages/sdk-ts/ TypeScript client SDK sdk-py/ Python client SDK benchmarks/browser-agent/ pre-registered browser-agent evaluation harness evals/ safety and durable crash-recovery evaluations scripts/ benchmark and release checks docs/research/ methods, findings and raw results examples/ runnable task examples MELRA is capability-driven, not language-restricted: Rust, Go, Swift, C or another language can be added when measurement shows a real benefit and the same conformance suite passes. Runnable examples https://github.com/XAGI-Lab/melra/blob/main/examples/README.md cover a read-only system task, a verified file write, a terminal check, browser inspection, scoped memory, computer capability discovery, a governed API call, independent verification, and a restart-safe workflow. Code, adapters, client integrations, benchmark harnesses, verifier predicates, documentation and threat analysis are all welcome. | Start here | What it is | |---|---| | good first issue https://github.com/XAGI-Lab/melra/labels/good%20first%20issue | Scoped, with the files to touch named in the issue | | help wanted https://github.com/XAGI-Lab/melra/labels/help%20wanted | Larger, still well defined | | Open design proposals https://github.com/XAGI-Lab/melra/issues?q=is%3Aissue+is%3Aopen+label%3Aproposal | Comment on open questions before implementation; see semantic verification 54 https://github.com/XAGI-Lab/melra/issues/54 | | Contributing guide https://github.com/XAGI-Lab/melra/blob/main/docs/CONTRIBUTING GUIDE.md | How the pipeline fits together and which files one change touches | 1. Read CONTRIBUTING.md https://github.com/XAGI-Lab/melra/blob/main/CONTRIBUTING.md and the Code of Conduct https://github.com/XAGI-Lab/melra/blob/main/CODE OF CONDUCT.md . 2. Sign your commits under the DCO git commit -s . 3. Run pnpm check before opening a pull request; it also checks links, the environment-variable table and the CHANGELOG shape. 4. Report vulnerabilities through private vulnerability reporting https://github.com/XAGI-Lab/melra/security/advisories/new , never in a public issue. Software and documentation are licensed under the Apache License 2.0 https://github.com/XAGI-Lab/melra/blob/main/LICENSE . The official logo and hero artwork are licensed under CC BY-ND 4.0 https://github.com/XAGI-Lab/melra/blob/main/LICENSES/CC-BY-ND-4.0.txt . Third-party benchmark datasets keep their own licenses and are not included in this repository. Built in the open by XAGI Labs https://github.com/XAGI-Lab . <sub MELRA brand assets © XAGI Labs Private Limited, licensed CC BY-ND 4.0.</sub