martin-loop 0.1.5 → 1.3.0

package/package.json

@@ -1,62 +1,134 @@
-{
-  "name": "martin-loop",
-  "private": false,
-  "version": "0.1.5",
-  "type": "module",
-  "description": "Martin Loop dual-track monorepo for the OSS runtime and hosted SaaS control plane.",
-  "license": "MIT",
-  "repository": {
-    "type": "git",
-    "url": "git+https://github.com/Keesan12/martin-loop.git"
-  },
-  "packageManager": "pnpm@10.17.1",
-  "main": "./dist/index.js",
-  "types": "./dist/index.d.ts",
-  "bin": {
-    "martin-loop": "./dist/bin/martin-loop.js",
-    "martin": "./dist/bin/martin-loop.js"
-  },
-  "exports": {
-    ".": {
-      "types": "./dist/index.d.ts",
-      "default": "./dist/index.js"
-    },
-    "./package.json": "./package.json"
-  },
+{
+  "name": "martin-loop",
+  "private": false,
+  "version": "1.3.0",
+  "type": "module",
+  "description": "Martin Loop dual-track monorepo for the OSS runtime and hosted SaaS control plane.",
+  "packageManager": "pnpm@10.33.0",
+  "keywords": [
+    "ai",
+    "agent",
+    "coding-loop",
+    "llm",
+    "automation",
+    "cli"
+  ],
+  "author": "Vakeesan Mahalingam and Gobi Shanthan",
+  "license": "MIT",
+  "homepage": "https://martinloop.com/",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/Keesan12/MartinLoop.git"
+  },
+  "main": "./dist/index.js",
+  "types": "./dist/index.d.ts",
+  "bin": {
+    "martin-loop": "./dist/bin/martin-loop.js"
+  },
+  "exports": {
+    ".": {
+      "types": "./dist/index.d.ts",
+      "default": "./dist/index.js"
+    },
+    "./package.json": "./package.json"
+  },
   "files": [
     "dist",
-    "demo",
     "README.md",
-    "docs/oss",
-    "docs/distribution"
+    "CODE_OF_CONDUCT.md",
+    "demo",
+    "docs/assets",
+    "docs/oss"
   ],
-  "publishConfig": {
-    "access": "public"
-  },
-  "scripts": {
-    "build": "pnpm -r build && node ./scripts/build-public-facade.mjs",
-    "prepublishOnly": "pnpm build",
-    "test": "pnpm -r test",
-    "lint": "pnpm -r lint",
-    "oss:validate": "node ./scripts/oss-boundary.mjs",
-    "release:surface:validate": "node ./scripts/release-surface-audit.mjs",
-    "pilot:prep:validate": "node ./scripts/pilot-prep-audit.mjs",
-    "public:smoke": "node ./scripts/public-facade-smoke.mjs",
-    "repo:smoke": "node ./scripts/repo-backed-smoke.mjs",
-    "rc:validate": "node ./scripts/rc-validation.mjs",
-    "rc:validate:install": "node ./scripts/rc-validation.mjs --install",
-    "release:matrix:local": "node ./scripts/release-matrix.mjs",
-    "dev:control-plane": "pnpm --filter @martin/control-plane dev",
-    "run:cli": "pnpm --filter @martin/cli dev",
-    "eval": "pnpm --filter @martin/benchmarks eval",
-    "dev:dashboard": "node apps/local-dashboard/server.js"
-  },
-  "devDependencies": {
-    "@types/node": "^22.13.10",
-    "@types/react": "^19.0.12",
-    "@types/react-dom": "^19.0.5",
-    "tsx": "^4.19.3",
-    "typescript": "^5.8.2",
-    "vitest": "^3.0.8"
-  }
-}
+  "publishConfig": {
+    "access": "public"
+  },
+  "bundleDependencies": [
+    "@martin/adapters",
+    "@martin/audit-exporter",
+    "@martin/cli",
+    "@martin/contracts",
+    "@martin/core",
+    "@martin/headlessos-core",
+    "@martin/policy",
+    "@martin/sdk",
+    "@martin/trace-intelligence"
+  ],
+  "scripts": {
+    "build": "pnpm -r build && node ./scripts/build-public-facade.mjs",
+    "test": "pnpm -r --filter=!@martin/audit-proof-tests test",
+    "test:audit-proof": "pnpm --filter @martin/audit-proof-tests test",
+    "lint": "pnpm -r lint",
+    "oss:validate": "node ./scripts/oss-boundary.mjs",
+    "agents:validate": "node ./scripts/validate-agent-discovery.mjs",
+    "agents:selftest": "node ./scripts/agent-readiness-selftest.mjs",
+    "db:verify-schema": "node ./scripts/db-verify-schema.mjs",
+    "release:surface:validate": "node ./scripts/release-surface-audit.mjs",
+    "pilot:prep:validate": "node ./scripts/pilot-prep-audit.mjs",
+    "pilot:intake:validate": "node ./scripts/pilot-intake-validate.mjs",
+    "pilot:live:preflight": "node ./scripts/pilot-live-preflight.mjs",
+    "pilot:review:validate": "node ./scripts/pilot-review-pack.mjs --self-check",
+    "pilot:gate:review": "node ./scripts/pilot-gate-review.mjs",
+    "public:smoke": "node ./scripts/public-facade-smoke.mjs",
+    "mcp:published:smoke": "pnpm --filter @martinloop/mcp smoke:published",
+    "examples:validate": "node ./scripts/examples-validate.mjs",
+    "repo:smoke": "node ./scripts/repo-backed-smoke.mjs",
+    "rc:validate": "node ./scripts/rc-validation.mjs",
+    "rc:validate:install": "node ./scripts/rc-validation.mjs --install",
+    "release:truth:validate": "node ./scripts/release-truth-audit.mjs",
+    "release:package:validate": "node ./scripts/release-package-validate.mjs",
+    "release:launch:validate": "node ./scripts/release-launch-readiness.mjs",
+    "release:gate:review": "node ./scripts/release-gate-review.mjs",
+    "release:scorecard": "node ./scripts/release-scorecard.mjs",
+    "release:audit:gaps": "node ./scripts/release-audit-gap-manifest.mjs",
+    "release:audit:handoff": "node ./scripts/package-external-audit-handoff.mjs",
+    "claims:lint": "node ./scripts/claim-lint.mjs",
+    "claims:compile": "node ./scripts/claim-compile.mjs",
+    "claims:trace-autonomy:gate": "node --import tsx/esm ./scripts/trace-autonomy-certification.ts --require-claim-allowed",
+    "claims:trace-autonomy:audit-request": "node --import tsx/esm ./scripts/trace-autonomy-certification.ts --write-audit-signoff-request .martin/runs/trace-autonomy-certification/audit-signoff-request.json",
+    "claims:trace-autonomy:run": "node --import tsx/esm ./scripts/trace-autonomy-certification-runner.ts",
+    "research:corpus:build": "node --import tsx/esm ./scripts/research-corpus/build.ts",
+    "research:corpus:report": "node --import tsx/esm ./scripts/research-corpus/report.ts",
+    "release:truth:compile": "node ./scripts/release-truth-compiler.mjs",
+    "release:matrix:local": "node ./scripts/release-matrix.mjs",
+    "dev:control-plane": "pnpm --filter @martin/control-plane dev",
+    "run:cli": "pnpm --filter @martin/cli dev",
+    "eval": "pnpm --filter @martin/benchmarks eval",
+    "eval:ralphy-stress": "pnpm --filter @martin/benchmarks eval:ralphy-stress",
+    "eval:known-bad-world": "node --import tsx/esm scripts/eval-known-bad-world.ts",
+    "dev:dashboard": "node apps/local-dashboard/server.js",
+    "post-phase15:handoff": "node ./scripts/package-post-phase15-handoff.mjs",
+    "eval:gate-check": "pnpm --filter @martin/sdk build && node --import tsx/esm scripts/eval-gate-check.ts",
+    "eval:case-studies": "node --import tsx/esm scripts/eval-case-studies.ts",
+    "eval:phase23": "pnpm --filter @martin/benchmarks phase23",
+    "security:deepsec:trial": "node --import tsx/esm ./scripts/security/deepsec-bounded-audit.ts",
+    "headless:worker:run": "node --import tsx/esm scripts/run-headless-worker.ts",
+    "headless:postgres:proof": "node --import tsx/esm scripts/headlessos-postgres-proof.ts",
+    "headless:live:proof": "node ./scripts/headlessos-live-proof.mjs",
+    "headless:handoff:package": "node ./scripts/package-headlessos-handoff.mjs"
+  },
+  "devDependencies": {
+    "@types/node": "^22.13.10",
+    "@types/pg": "^8.20.0",
+    "@types/react": "^19.0.12",
+    "@types/react-dom": "^19.0.5",
+    "pg": "^8.20.0",
+    "tsx": "^4.19.3",
+    "typescript": "^5.8.2",
+    "vitest": "^3.0.8"
+  },
+  "dependencies": {
+    "@open-policy-agent/opa-wasm": "^1.10.0",
+    "@opentelemetry/api-logs": "^0.214.0",
+    "@opentelemetry/exporter-logs-otlp-http": "^0.214.0",
+    "@opentelemetry/resources": "^2.6.1",
+    "@opentelemetry/sdk-logs": "^0.214.0",
+    "@opentelemetry/semantic-conventions": "^1.38.0",
+    "ts-morph": "^21.0.0"
+  },
+  "workspaces": [
+    "packages/*",
+    "apps/*",
+    "benchmarks"
+  ]
+}

package/docs/distribution/DIRECTORY-SUBMISSIONS.md

@@ -1,89 +0,0 @@
-# Directory Submission Pack
-
-Use this file as the single source of truth for public directory submissions.
-
-## Short tagline
-
-Open-source control plane for AI coding agents.
-
-## Long description
-
-MartinLoop is an open-source governed runtime for AI coding agents. It wraps autonomous coding loops with budget caps, verifier gates, rollback evidence, JSONL run records, failure classification, and MCP/Claude/Codex integration so agent work can be inspected, halted, and trusted.
-
-## Primary links
-
-- GitHub repo: [github.com/Keesan12/martin-loop](https://github.com/Keesan12/martin-loop)
-- Website: [martinloop.com](https://martinloop.com)
-- npm package: [npmjs.com/package/martin-loop](https://www.npmjs.com/package/martin-loop)
-- Benchmark challenge: [UNDER-3-CHALLENGE.md](./UNDER-3-CHALLENGE.md)
-
-## Submission checklist
-
-### OpenAlternative
-
-- status: pending
-- surface: OSS alternative listing
-- copy to use: short tagline + long description
-- include: GitHub, website, npm
-
-### DevHunt
-
-- status: pending
-- surface: product hunt style dev tools directory
-- copy to use: short tagline + long description
-- include: benchmark challenge and demo command
-
-### Uneed
-
-- status: pending
-- surface: startup/tool discovery
-- copy to use: short tagline + long description
-- include: GitHub, website, npm
-
-### BetaList
-
-- status: pending
-- surface: early product discovery
-- copy to use: short tagline + long description
-- include: why governed agent runs matter
-
-### Microlaunch
-
-- status: pending
-- surface: lightweight launch directory
-- copy to use: short tagline + long description
-- include: demo command and benchmark challenge
-
-### AlternativeTo
-
-- status: pending
-- surface: alternative comparison listing
-- copy to use: short tagline + long description
-- include: comparable tools and differentiators
-
-### Futurepedia
-
-- status: pending
-- surface: AI tools directory
-- copy to use: short tagline + long description
-- include: Claude, Codex, and MCP integration
-
-### Toolify
-
-- status: pending
-- surface: AI tool directory
-- copy to use: short tagline + long description
-- include: benchmark challenge link
-
-### There’s An AI For That
-
-- status: pending
-- surface: AI tool catalog
-- copy to use: short tagline + long description
-- include: GitHub, website, npm
-
-## Notes
-
-- Prefer submissions that link directly to the repo, website, and npm package together.
-- Reuse the benchmark challenge and `martin-loop demo` as the fastest trust-building assets.
-- If a directory wants screenshots, use the current public repo README visuals instead of inventing a separate pitch deck.

package/docs/distribution/INTEGRATION-OUTREACH.md

@@ -1,61 +0,0 @@
-# Integration Outreach Pack
-
-Use this file for direct outreach to projects and communities building around AI coding agents.
-
-## Core message
-
-Hey [Name] — I’m building MartinLoop, an OSS governed runtime for AI coding agents.
-
-The repo already supports budget caps, verifier gates, JSONL run records, rollback evidence, Claude/Codex adapters, and an MCP package.
-
-I’m trying to understand where a control layer like this should integrate best with projects like [their project]: CLI wrapper, MCP boundary, CI, or runtime adapter.
-
-Would value your blunt take — useful direction or wrong abstraction?
-
-## Target projects
-
-- Claude Code
-- Codex CLI
-- MCP servers
-- Aider
-- Cline
-- Continue
-- OpenHands
-- SWE-agent
-- Goose
-- DevContainers
-- GitHub Actions
-
-## Outreach notes by target
-
-### Claude Code
-
-- emphasize governed repo runs and MCP install path
-- ask whether the best control point is local CLI wrapper or MCP boundary
-
-### Codex CLI
-
-- emphasize budget caps, verifier gates, and auditable run records
-- ask whether wrapper, runtime adapter, or CI integration is most useful
-
-### MCP projects
-
-- emphasize the packaged `@martinloop/mcp` server surface
-- ask whether the trust layer belongs at tool boundary or runtime boundary
-
-### Aider, Cline, Continue, OpenHands, SWE-agent, Goose
-
-- emphasize adapter-normalized receipts and halt reasons
-- ask how much control should live in the agent runtime versus CI or wrapper
-
-### DevContainers and GitHub Actions
-
-- emphasize safe default automation, budget visibility, and verifier gates in shared team workflows
-- ask where platform teams want policy to live
-
-## Supporting assets
-
-- challenge page: [UNDER-3-CHALLENGE.md](./UNDER-3-CHALLENGE.md)
-- directory copy: [DIRECTORY-SUBMISSIONS.md](./DIRECTORY-SUBMISSIONS.md)
-- repo: [github.com/Keesan12/martin-loop](https://github.com/Keesan12/martin-loop)
-- npm: [npmjs.com/package/martin-loop](https://www.npmjs.com/package/martin-loop)

package/docs/distribution/UNDER-3-CHALLENGE.md

@@ -1,65 +0,0 @@
-# Can your AI coding agent finish this task under $3?
-
-MartinLoop is testing a simple question:
-
-Can an AI coding agent complete a task under a fixed budget, with verifier-passed completion and an inspectable run record?
-
-## Current repo-backed comparison
-
-Same task, same starting state:
-
-- governed MartinLoop run: `$2.30`
-- uncontrolled retry loop: `$5.20`
-- governed outcome: `completed` and verifier-passed with an inspectable record
-- uncontrolled outcome: failed after repeated retries with no comparable audit trail
-
-These numbers match the current public benchmark story shown in the repo README and visualized in [`docs/assets/side-by-side.svg`](../assets/side-by-side.svg).
-
-## Why this matters
-
-The claim is not that every governed run is always cheaper. The claim is that the run becomes inspectable and enforceable:
-
-- budget policy is explicit
-- verifier success is explicit
-- stop reasons are explicit
-- artifacts are inspectable after the run
-
-That makes a coding-agent result easier to trust, replay, compare, and audit.
-
-## Reproduce it
-
-From the repo root:
-
-```bash
-pnpm --filter @martin/benchmarks test
-pnpm --filter @martin/benchmarks eval
-pnpm --filter @martin/benchmarks eval:phase12
-```
-
-## What to share back
-
-If you run a similar challenge with Claude Code, Codex CLI, Cursor, Aider, Cline, Continue, OpenHands, SWE-agent, Goose, or an internal coding agent, share:
-
-- total budget used
-- number of attempts
-- verifier result
-- whether the final run was auditable
-- whether rollback evidence was available
-
-## Try MartinLoop without risking your repo
-
-You can copy the public demo sandbox first:
-
-```bash
-npx martin-loop demo
-```
-
-Then run the sandbox locally with the printed next steps.
-
-## Claim boundary
-
-This page intentionally stays inside the current public evidence boundary:
-
-- the `$2.30` and `$5.20` figures are the current repo-backed benchmark story used in the public README
-- the reproduction commands above are real commands from this repository
-- the benchmark harness remains a workspace-level surface, so challenge claims should stay tied to repo-backed outputs rather than generic marketing numbers