{
  "markdown": "# Laviya Agent Skills\n\nLaviya Agent Skills is a repository of reusable orchestration assets for IDE and agent integrations.\nIt provides a production-oriented MCP runtime, shared orchestration prompt assets, and client-specific\nartifacts for tools such as Cursor and Claude.\n\n## Packages\n\n- `laviya-agent-skills`: reusable prompt, rule, and skill asset bundle for Node-based tooling\n- `laviya-mcp-server`: stdio MCP runtime published from `mcp/`\n\n## What Is Included\n\n- `core/`: shared orchestration prompt assets\n- `mcp/`: TypeScript MCP runtime scaffold (`laviya-mcp-server`)\n- `cursor/`: Cursor rule artifacts\n- `claude/`: Claude skill artifacts\n- `docs/`: setup, npm publishing, and end-user onboarding guides\n\n## Repository Structure\n\n```text\nlaviya-agent-skills/\n  core/\n  mcp/\n    src/\n    examples/\n    README.md\n  cursor/\n  claude/\n  docs/\n    Setup.md\n    MCPServerPublish.md\n    InstallationAndUsage.md\n```\n\n## Quick Start\n\n```bash\ncd mcp\nnpm install\nnpm run typecheck\nnpm run build\nnpm run dev\n```\n\nRequired environment variable:\n\n- `LAVIYA_API_KEY`\n\n## Installable Asset Package\n\nInstall the reusable asset bundle:\n\n```bash\nnpm install laviya-agent-skills\n```\n\nExample usage:\n\n```js\nimport { assets, resolveAssetPath } from \"laviya-agent-skills\";\nimport { readFileSync } from \"node:fs\";\n\nconst cursorRule = readFileSync(assets.cursorRule, \"utf8\");\nconst orchestratorPrompt = readFileSync(resolveAssetPath(\"orchestratorSystemPrompt\"), \"utf8\");\n```\n\n## Documentation Index\n\n- Runtime architecture and implementation: `mcp/README.md`\n- Developer setup guide: `docs/Setup.md`\n- NPM publishing guide: `docs/MCPServerPublish.md`\n- End-user installation and usage: `docs/InstallationAndUsage.md`\n- Client-specific MCP setup (Codex, VS Code, Antigravity, Claude): `docs/InstallationAndUsage.md`\n\n## Runtime Overview\n\nThe MCP runtime follows a global-runtime plus project-local-config model:\n\n- machine-level runtime installation\n- repository-level project configuration (`.laviya/project.json` or `.laviya.json`)\n- runtime-enforced lifecycle mechanics (validation, retries, idempotency, and structured logging)\n\nCurrent MCP tool set includes both orchestration and local-direct task feed flows:\n\n- `laviya_help`\n- `laviya_feed_task`\n- `laviya_get_local_work_status`\n- `laviya_cancel_local_work`\n- `laviya_add_task_comment`\n- `laviya_get_my_work`\n- `laviya_start_execution`\n- `laviya_complete_execution`\n- `laviya_report_token_usage`\n- `laviya_diagnostics`\n\nAll tools return API envelope JSON text in the shape: `{ HasFailed, Messages, Data }`.\n\nWork items may include a backend-defined `ExecutionPolicy`. Analysis and review\npolicies are read-only: agents must not implement changes, and enforced policies\nrequire matching completion `executionEvidence` and summary `policyCompliance`.\n\nFor full architecture details and file-level scaffolding, refer to `mcp/README.md`.\n",
  "bytes": 2875,
  "sha": "97742ec74a9f619e9fb4b64b1de826ce40a76bba5a30be3f1c5af0701f268c6c",
  "repo_slug": "kartegagithub/laviya-agent-skills",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_kartegagithub_laviya_agent_ski_a05eb9ca/readme"
}