{"id":24911270,"url":"https://github.com/kdeps/kdeps","last_synced_at":"2026-05-16T07:15:55.115Z","repository":{"id":272205743,"uuid":"836501717","full_name":"kdeps/kdeps","owner":"kdeps","description":"Straightforward LLM dependency orchestration for multi-agent workflows. Compose chat, code, and data into declarative pipelines in YAML. Export AI workflows as a single binary, ISO, Docker or Kubernetes pods. Use ollama, llmfile, or any cloud AI providers.","archived":false,"fork":false,"pushed_at":"2026-05-09T07:12:41.000Z","size":80695,"stargazers_count":32,"open_issues_count":13,"forks_count":4,"subscribers_count":1,"default_branch":"main","last_synced_at":"2026-05-09T07:36:11.600Z","etag":null,"topics":["agent","llms","orchestration","workflow","yaml"],"latest_commit_sha":null,"homepage":"https://kdeps.com","language":"Go","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/kdeps.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":"NOTICE","maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2024-08-01T01:46:41.000Z","updated_at":"2026-05-09T07:12:45.000Z","dependencies_parsed_at":"2026-04-19T07:02:10.127Z","dependency_job_id":null,"html_url":"https://github.com/kdeps/kdeps","commit_stats":null,"previous_names":["kdeps/kdeps"],"tags_count":157,"template":false,"template_full_name":null,"purl":"pkg:github/kdeps/kdeps","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kdeps%2Fkdeps","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kdeps%2Fkdeps/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kdeps%2Fkdeps/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kdeps%2Fkdeps/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/kdeps","download_url":"https://codeload.github.com/kdeps/kdeps/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kdeps%2Fkdeps/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33093885,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-16T04:41:52.686Z","status":"ssl_error","status_checked_at":"2026-05-16T04:41:52.009Z","response_time":115,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agent","llms","orchestration","workflow","yaml"],"created_at":"2025-02-02T04:19:02.166Z","updated_at":"2026-05-16T07:15:55.087Z","avatar_url":"https://github.com/kdeps.png","language":"Go","funding_links":[],"categories":[],"sub_categories":[],"readme":"# kdeps\n\n**Straightforward LLM dependency orchestration for multi-agent workflows.** Compose chat, code, and data into declarative pipelines in YAML. Export AI workflows as a single binary, ISO, Docker, or Kubernetes pods. Use Ollama, llamafile, or any cloud AI provider.\n\n\u003e **Highly experimental.** APIs, YAML schemas, and CLI flags can change without notice. Do not use in production. [Report issues or give feedback](https://github.com/kdeps/kdeps/issues).\n\n## Why kdeps?\n\nChat AIs and their MCP extensions are tools you operate. kdeps is for building **deployable AI workflows** — pipelines that chain LLM calls with code execution, data lookups, and API requests, then export as a binary, Docker, ISO, or Kubernetes pod.\n\n| | Chat AI + MCP | kdeps |\n|---|---|---|\n| **Deployed as** | A chat session | Binary, Docker, ISO, Kubernetes |\n| **Logic lives in** | Prompts | YAML — versioned, reviewed, tested |\n| **Orchestration** | Model-driven | Explicit dependency pipelines |\n| **Ships to production** | No | Yes |\n\n## Install\n\n```bash\ncurl -LsSf https://raw.githubusercontent.com/kdeps/kdeps/main/install.sh | sh\n```\n\n## Quick start\n\n```bash\nkdeps new                    # scaffold a project\nkdeps run workflow.yaml --dev  # hot-reload, no Docker needed\n```\n\nA minimal agent that answers questions via an LLM:\n\n```yaml\n# resources/chat.yaml\napiVersion: kdeps.io/v1\nkind: Resource\nmetadata:\n  actionId: chat\nrun:\n  chat:\n    prompt: \"{{ get('message') }}\"\n  apiResponse:\n    response: \"{{ output('chat') }}\"\n```\n\nWire resources into pipelines — outputs flow between steps via `requires:`:\n\n```yaml\nmetadata:\n  actionId: fetch\nrun:\n  scraper:\n    url: \"{{ get('url') }}\"\n\n---\nmetadata:\n  actionId: summarize\n  requires: [fetch]\nrun:\n  chat:\n    prompt: \"Summarize: {{ output('fetch').content }}\"\n  apiResponse:\n    response: \"{{ output('summarize') }}\"\n```\n\n## Build and deploy\n\n```bash\nkdeps bundle build          # Docker image\nkdeps export k8s            # Kubernetes manifests\nkdeps bundle export iso     # bootable edge ISO\nkdeps bundle prepackage     # self-contained binary per arch\n```\n\n## Global config\n\n```bash\nkdeps edit    # opens ~/.kdeps/config.yaml\nkdeps doctor  # check system health (config, Ollama, Python, agents)\n```\n\nConfig is validated on load. Warnings are printed to stderr for:\n- Typos in API key / field names\n- Backend set without a corresponding API key\n- Invalid routing strategy values\n- Malformed duration strings\n- Agent profiles not matching any installed workflow\n- Empty agent profiles\n\nWarnings and errors use structured JSON logging (via `log/slog`). Set `KDEPS_LOG_FORMAT=json` for production JSON output. Log level defaults to WARN; use `--verbose` for INFO or `--debug` for DEBUG.\n\n```yaml\nllm:\n  backend: ollama           # ollama, openai, anthropic, groq, ...\n  # openai_api_key: sk-...\n  # anthropic_api_key: sk-ant-...\n\ndefaults:\n  timezone: UTC\n  python_version: \"3.12\"\n\nresource_defaults:\n  chat:\n    timeout: \"60s\"\n    context_length: 4096\n  http:\n    timeout: \"30s\"\n```\n\n**Security** - set in `workflow.yaml` under `settings.apiServer`: `auth.token` requires a Bearer or `X-Api-Key` header on every request; `rateLimit` enforces per-IP throttling; `maxBodyBytes` caps request body size. Enable TLS via `settings.certFile` / `settings.keyFile`. See [Security](https://kdeps.com/configuration/advanced#security).\n\n---\n\n[Documentation](https://kdeps.com) | [Registry](https://kdeps.io) | Apache 2.0\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkdeps%2Fkdeps","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkdeps%2Fkdeps","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkdeps%2Fkdeps/lists"}