agentloopkit 0.1.0

package/dist/templates/root/AGENTLOOP.md

@@ -0,0 +1,39 @@
+# AgentLoopKit
+
+AgentLoopKit gives coding agents a repeatable engineering loop inside this repo.
+
+## Loop
+
+1. Specify: turn the request into a task contract with outcome, constraints, non-goals, likely files, and files not to touch.
+2. Constrain: define safety rules, dependency rules, verification commands, and rollback notes.
+3. Plan: inspect relevant files, identify risks, and write a short implementation plan.
+4. Implement: make focused changes and avoid unrelated formatting churn.
+5. Verify: run configured checks and capture pass or fail results.
+6. Review: inspect the diff against the task contract and identify risks.
+7. Handoff: summarize changed files, verification, risks, rollback notes, and reviewer checklist.
+
+## Repo Layout
+
+- .agentloop/loops/: workflow templates for different task types.
+- .agentloop/gates/: checklists for implementation, tests, review, regression, docs, security, and dependencies.
+- .agentloop/policies/: practical safety policies for agents.
+- .agentloop/tasks/: task contracts for agentic coding sessions.
+- .agentloop/reports/: verification reports generated by `agentloop verify`.
+- .agentloop/handoffs/: PR summaries and reviewer handoffs.
+- .agentloop/harness/: repo-specific working agreement, commands, and checklists.
+
+## Commands Detected During Init
+
+- Package manager: {{ packageManager }}
+- Project type: {{ projectType }}
+- Test: {{ testCommand }}
+- Lint: {{ lintCommand }}
+- Typecheck: {{ typecheckCommand }}
+- Build: {{ buildCommand }}
+- Format: {{ formatCommand }}
+
+## Autonomous Work
+
+Agents may work autonomously inside the task contract. They should stop and ask before changing protected areas, adding dependencies, introducing breaking public APIs, or taking destructive actions.
+
+No agent should claim completion without verification evidence or a clear statement of what was not verified.

package/dist/templates/root/AGENTS.md

@@ -0,0 +1,24 @@
+# AGENTS
+
+<!-- agentloopkit:start -->
+
+This repository uses AgentLoopKit.
+
+Before changing code:
+
+- Read AGENTLOOP.md.
+- Check .agentloop/tasks/ for an active task contract.
+- Follow the Specify, Constrain, Plan, Implement, Verify, Review, Handoff loop.
+- Keep changes small and tied to the task contract.
+- Do not run destructive git or filesystem commands unless the user asks for them.
+- Do not read or print secrets. If env files exist, mention only their paths.
+- Run the configured verification commands before claiming completion.
+- Generate a handoff summary with changed files, tests run, risks, and rollback notes.
+
+Safety rules:
+
+- Treat migrations, auth, billing, deployment, lockfiles, public APIs, and security code as high-risk.
+- Do not change dependencies without a clear reason and verification plan.
+- Preserve existing user work. Do not revert unrelated changes.
+- Update DECISIONS.md when architecture or public behavior changes.
+<!-- agentloopkit:end -->

package/dist/templates/root/agentloop-directory-readme.md

@@ -0,0 +1,36 @@
+# AgentLoopKit Workspace
+
+This directory contains repo-local engineering loop artifacts for coding agents and reviewers.
+
+## Start Here
+
+1. Read `../AGENTS.md`.
+2. Read `../AGENTLOOP.md`.
+3. Create or inspect a task contract:
+
+```bash
+agentloop create-task --title "Describe the next focused change" --type feature
+```
+
+4. Run verification when work is ready:
+
+```bash
+agentloop verify
+```
+
+5. Generate a reviewer handoff:
+
+```bash
+agentloop summarize --write
+```
+
+## Directories
+
+- `loops/`: task-specific workflows
+- `gates/`: pass/fail checklists
+- `policies/`: safety rules
+- `tasks/`: task contracts
+- `reports/`: verification reports
+- `handoffs/`: PR summaries and reviewer briefs
+- `agents/`: agent-specific instructions
+- `harness/`: repo working agreement and commands

package/dist/templates/root/agentloop.config.json

@@ -0,0 +1,37 @@
+{
+  "$schema": "https://agentloopkit.dev/schema/agentloop.config.schema.json",
+  "version": 1,
+  "project": {
+    "name": "{{ projectName }}",
+    "type": "{{ projectType }}",
+    "packageManager": "{{ packageManager }}"
+  },
+  "paths": {
+    "root": ".",
+    "agentloopDir": ".agentloop",
+    "tasksDir": ".agentloop/tasks",
+    "reportsDir": ".agentloop/reports",
+    "handoffsDir": ".agentloop/handoffs"
+  },
+  "commands": {
+    "test": "",
+    "lint": "",
+    "typecheck": "",
+    "build": "",
+    "format": ""
+  },
+  "safety": {
+    "requireCleanWorkingTree": false,
+    "warnOnDirtyWorkingTree": true,
+    "protectEnvFiles": true,
+    "protectMigrations": true,
+    "protectLockfiles": false
+  },
+  "summary": {
+    "includeDiffStats": true,
+    "includeChangedFiles": true,
+    "includeVerification": true,
+    "includeRisks": true,
+    "includeRollback": true
+  }
+}

package/dist/templates/tasks/README.md

@@ -0,0 +1,9 @@
+# Task Contracts
+
+Create task contracts with:
+
+```bash
+agentloop create-task
+```
+
+Task contracts turn fuzzy requests into scoped engineering work. A good contract names the desired outcome, constraints, non-goals, likely files, files not to touch, acceptance criteria, verification commands, and rollback notes.

package/package.json

@@ -0,0 +1,71 @@
+{
+  "name": "agentloopkit",
+  "version": "0.1.0",
+  "description": "A drop-in engineering loop for coding agents.",
+  "type": "module",
+  "bin": {
+    "agentloop": "dist/cli/index.js",
+    "agentloopkit": "dist/cli/index.js"
+  },
+  "files": [
+    "dist",
+    "schema",
+    "README.md",
+    "LICENSE"
+  ],
+  "keywords": [
+    "ai",
+    "agents",
+    "coding-agents",
+    "codex",
+    "claude-code",
+    "cursor",
+    "opencode",
+    "gemini-cli",
+    "agentic-engineering",
+    "vibe-coding",
+    "developer-tools",
+    "cli"
+  ],
+  "license": "MIT",
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/abhiyoheswaran1/AgentLoopKit.git"
+  },
+  "homepage": "https://github.com/abhiyoheswaran1/AgentLoopKit#readme",
+  "bugs": {
+    "url": "https://github.com/abhiyoheswaran1/AgentLoopKit/issues"
+  },
+  "engines": {
+    "node": ">=20"
+  },
+  "packageManager": "pnpm@10.12.1",
+  "scripts": {
+    "build": "tsup && node scripts/copy-assets.mjs",
+    "dev": "tsx src/cli/index.ts",
+    "test": "vitest run",
+    "test:watch": "vitest",
+    "typecheck": "tsc --noEmit",
+    "lint": "eslint .",
+    "format": "prettier --write .",
+    "prepublishOnly": "npm run typecheck && npm test && npm run build"
+  },
+  "dependencies": {
+    "commander": "^14.0.2",
+    "execa": "^9.6.0",
+    "prompts": "^2.4.2",
+    "zod": "^4.1.13"
+  },
+  "devDependencies": {
+    "@eslint/js": "^10.0.1",
+    "@types/node": "^25.0.0",
+    "@types/prompts": "^2.4.9",
+    "eslint": "^10.4.1",
+    "prettier": "^3.7.4",
+    "tsup": "^8.5.1",
+    "tsx": "^4.21.0",
+    "typescript": "^5.9.3",
+    "typescript-eslint": "^8.61.0",
+    "vitest": "^4.0.15"
+  }
+}

package/schema/agentloop.config.schema.json

@@ -0,0 +1,142 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://agentloopkit.dev/schema/agentloop.config.schema.json",
+  "title": "AgentLoopKit Configuration",
+  "type": "object",
+  "additionalProperties": false,
+  "required": ["version", "project", "paths", "commands", "safety", "summary"],
+  "properties": {
+    "$schema": {
+      "type": "string"
+    },
+    "version": {
+      "const": 1
+    },
+    "project": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": ["name", "type", "packageManager"],
+      "properties": {
+        "name": {
+          "type": "string"
+        },
+        "type": {
+          "type": "string",
+          "enum": [
+            "nextjs",
+            "react-vite",
+            "node",
+            "typescript-package",
+            "python",
+            "docs-only",
+            "generic"
+          ]
+        },
+        "packageManager": {
+          "type": "string",
+          "enum": ["pnpm", "npm", "yarn", "bun"]
+        }
+      }
+    },
+    "paths": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": ["root", "agentloopDir", "tasksDir", "reportsDir", "handoffsDir"],
+      "properties": {
+        "root": {
+          "type": "string"
+        },
+        "agentloopDir": {
+          "type": "string"
+        },
+        "tasksDir": {
+          "type": "string"
+        },
+        "reportsDir": {
+          "type": "string"
+        },
+        "handoffsDir": {
+          "type": "string"
+        }
+      }
+    },
+    "commands": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": ["test", "lint", "typecheck", "build", "format"],
+      "properties": {
+        "test": {
+          "type": "string"
+        },
+        "lint": {
+          "type": "string"
+        },
+        "typecheck": {
+          "type": "string"
+        },
+        "build": {
+          "type": "string"
+        },
+        "format": {
+          "type": "string"
+        }
+      }
+    },
+    "safety": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [
+        "requireCleanWorkingTree",
+        "warnOnDirtyWorkingTree",
+        "protectEnvFiles",
+        "protectMigrations",
+        "protectLockfiles"
+      ],
+      "properties": {
+        "requireCleanWorkingTree": {
+          "type": "boolean"
+        },
+        "warnOnDirtyWorkingTree": {
+          "type": "boolean"
+        },
+        "protectEnvFiles": {
+          "type": "boolean"
+        },
+        "protectMigrations": {
+          "type": "boolean"
+        },
+        "protectLockfiles": {
+          "type": "boolean"
+        }
+      }
+    },
+    "summary": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [
+        "includeDiffStats",
+        "includeChangedFiles",
+        "includeVerification",
+        "includeRisks",
+        "includeRollback"
+      ],
+      "properties": {
+        "includeDiffStats": {
+          "type": "boolean"
+        },
+        "includeChangedFiles": {
+          "type": "boolean"
+        },
+        "includeVerification": {
+          "type": "boolean"
+        },
+        "includeRisks": {
+          "type": "boolean"
+        },
+        "includeRollback": {
+          "type": "boolean"
+        }
+      }
+    }
+  }
+}