--- title: "agent-starter-pack vs kitaru" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/googlecloudplatform-agent-starter-pack-vs-zenml-io-kitaru" tools: ["googlecloudplatform-agent-starter-pack", "zenml-io-kitaru"] --- # agent-starter-pack vs kitaru *GraphCanon updated Aug 20, 2026* ## Verdict Pick agent-starter-pack if agent-starter-pack offers built-in CI/CD, evaluation tools, and observability for deploying generative AI agents on Google Cloud with Python; pick kitaru if kitaru focuses on recording, replaying, and enhancing the performance of AI agents in production environments using technology from ZenML. [agent-starter-pack](http://goo.gle/agents-cli) reports 6.5k GitHub stars, 1.5k forks, and 49 open issues, last pushed Jul 21, 2026. [kitaru](https://kitaru.ai) has 226 stars, 15 forks, and 49 open issues, last pushed Aug 3, 2026. Figures are from public GitHub metadata via [agent-starter-pack's repository](https://github.com/GoogleCloudPlatform/agent-starter-pack) and [kitaru's repository](https://github.com/zenml-io/kitaru). | | [agent-starter-pack](/tools/googlecloudplatform-agent-starter-pack.md) | [kitaru](/tools/zenml-io-kitaru.md) | | --- | --- | --- | | Tagline | Ship AI Agents to Google Cloud in minutes with built-in CI/CD, evaluation, and observability | Record, replay, and improve AI agents in production, built on ZenML | | Stars | 6,537 | 226 | | Forks | 1,502 | 15 | | Open issues | 49 | 49 | | Language | Python | Python | | Adopt for | agent-starter-pack offers built-in CI/CD, evaluation tools, and observability for deploying generative AI agents on Google Cloud with Python. | Kitaru focuses on recording, replaying, and enhancing the performance of AI agents in production environments using technology from ZenML. | | Persona | - | - | | Runtime | - | - | | License | Apache-2.0 | Apache-2.0 | | Categories | AI Agents, Evaluation & Observability | AI Agents, Evaluation & Observability | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [agent-starter-pack](/tools/googlecloudplatform-agent-starter-pack.md) | [kitaru](/tools/zenml-io-kitaru.md) | | --- | --- | --- | | Maintenance | Active (82%) | Very active (96%) | | Days since push | 29d | 0d | | Stars delta | +18 (30d) | Unknown | | Open issues delta | +1 (30d) | Unknown | | Full report | [trust report](/tools/googlecloudplatform-agent-starter-pack/trust.md) | [trust report](/tools/zenml-io-kitaru/trust.md) | ## Shared compatibility - **Python**: [agent-starter-pack](/tools/googlecloudplatform-agent-starter-pack.md) - Python runtime; [kitaru](/tools/zenml-io-kitaru.md) - Python runtime ## Decision facts: agent-starter-pack - **Requirements:** Depends on Python 3.10+.; Requires the Google Cloud SDK for interacting with GCP services.; Terraform is needed for deployment purposes.; Make utility should be installed for development tasks. - **Adopt for:** agent-starter-pack offers built-in CI/CD, evaluation tools, and observability for deploying generative AI agents on Google Cloud with Python. ## Decision facts: kitaru - **Adopt for:** Kitaru focuses on recording, replaying, and enhancing the performance of AI agents in production environments using technology from ZenML. ## Choose when ### Choose agent-starter-pack if… - Requirements: Depends on Python 3.10+.; Requires the Google Cloud SDK for interacting with GCP services.; Terraform is needed for deployment purposes.; Make utility should be installed for development tasks.. - Tags unique to agent-starter-pack: agents, gcp, gemini, genai-agents. - You are working with Google Cloud and wish to deploy generative AI agents quickly using production-ready templates that come with integrated CI/CD pipelines. ### Choose kitaru if… - Tags unique to kitaru: agent-framework, ai-agents, checkpoints, durable-execution. - - You need to ensure the continuous improvement of AI agents that are already deployed; Kitaru allows you to replay scenarios with different approaches to identify improvements. - More recently updated (last pushed Aug 3, 2026). ## When NOT to use agent-starter-pack - Your project is hosted on a cloud provider other than Google Cloud, as this tool is optimized for GCP services and uses Terraform specifically configured for deployment with it. - You do not need or want built-in CI/CD pipelines and evaluation tools, preferring to manage these components separately through different tools or self-configured setups. ## When NOT to use kitaru - - If your project is in the early stages of development without a clear need for replaying historical data or improving upon past behaviors; - - When working outside Python, as Kitaru does not currently offer support for other programming languages. ## Common questions ### What is the difference between agent-starter-pack and kitaru? agent-starter-pack: Ship AI Agents to Google Cloud in minutes with built-in CI/CD, evaluation, and observability. kitaru: Record, replay, and improve AI agents in production, built on ZenML. See the comparison table for live GitHub stats and shared categories. ### When should I choose agent-starter-pack over kitaru? Choose agent-starter-pack over kitaru when Requirements: Depends on Python 3.10+.; Requires the Google Cloud SDK for interacting with GCP services.; Terraform is needed for deployment purposes.; Make utility should be installed for development tasks.; Tags unique to agent-starter-pack: agents, gcp, gemini, genai-agents; You are working with Google Cloud and wish to deploy generative AI agents quickly using production-ready templates that come with integrated CI/CD pipelines. ### When should I choose kitaru over agent-starter-pack? Choose kitaru over agent-starter-pack when Tags unique to kitaru: agent-framework, ai-agents, checkpoints, durable-execution; - You need to ensure the continuous improvement of AI agents that are already deployed; Kitaru allows you to replay scenarios with different approaches to identify improvements; More recently updated (last pushed Aug 3, 2026). ### When should I avoid agent-starter-pack? Your project is hosted on a cloud provider other than Google Cloud, as this tool is optimized for GCP services and uses Terraform specifically configured for deployment with it. You do not need or want built-in CI/CD pipelines and evaluation tools, preferring to manage these components separately through different tools or self-configured setups. ### When should I avoid kitaru? - If your project is in the early stages of development without a clear need for replaying historical data or improving upon past behaviors; - When working outside Python, as Kitaru does not currently offer support for other programming languages. ### Is agent-starter-pack or kitaru more popular on GitHub? agent-starter-pack has more GitHub stars (6,537 vs 226). Stars measure visibility, not whether either tool fits your constraints. ### Are agent-starter-pack and kitaru open source? Yes - both are open-source projects on GitHub (agent-starter-pack: Apache-2.0, kitaru: Apache-2.0). ### Where can I find alternatives to agent-starter-pack or kitaru? GraphCanon lists graph-backed alternatives at [agent-starter-pack alternatives](/tools/googlecloudplatform-agent-starter-pack/alternatives) and [kitaru alternatives](/tools/zenml-io-kitaru/alternatives) ([agent-starter-pack markdown twin](/tools/googlecloudplatform-agent-starter-pack/alternatives.md), [kitaru markdown twin](/tools/zenml-io-kitaru/alternatives.md)), ranked by typed relationship edges rather than popularity votes. ### Is there a machine-readable version of this comparison? Yes. The markdown twin at [this comparison](/compare/googlecloudplatform-agent-starter-pack-vs-zenml-io-kitaru.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, agent-starter-pack or kitaru? agent-starter-pack: Active. kitaru: Very active. Compare maintenance labels, days since push, and release cadence in the trust section below - stars alone do not measure maintenance. ### Where are the full trust reports for agent-starter-pack and kitaru? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [agent-starter-pack trust report](/tools/googlecloudplatform-agent-starter-pack/trust); [kitaru trust report](/tools/zenml-io-kitaru/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=googlecloudplatform-agent-starter-pack`](/api/graphcanon/graph?tool=googlecloudplatform-agent-starter-pack) - LLM index: [/llms.txt](/llms.txt) - Full corpus: [/llms-full.txt](/llms-full.txt) _GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_