# Kiro Crew Persistent bots for development work: with individual memory, schedules and coordinated worker sessions on your computer or server. Canonical profile: https://ki-agentenkompass.de/en/agents/kiro-crew/ Editor: [Michael Meierhoff](https://ki-agentenkompass.de/en/about/) Sources reviewed: 2026-11-06 Basis: provider documentation. Examples are editorial suggestions, not hands-on product tests. ## Overview Kiro Crew belongs to the Kiro family and normally uses Kiro CLI through ACP. Crew adds persistent bots and orchestration; the Kiro profile covers the shared agent core and its other interfaces. - Provider: AWS - Product family: Kiro - Profile type: Persistent bot workspace - Environment: Desktop app, web dashboard and CLI; your own computer or server - Connections: MCP, skills, Slack, Discord, Telegram and other chat channels - Setup: Install Crew, sign in to Kiro CLI and configure a bot with its task, tools and approvals. - Pricing model: Apache-2.0 software; Kiro credits and your own hosting ## Key features ### Bots with individual responsibilities Named bots retain goals, a continuing chat and their own workspace. Source: https://kiro.dev/blog/crew-bots/ ### Separate bot memory Each bot has its own memories that you can inspect, edit and delete. Source: https://kiro.dev/blog/crew-bots/ ### Coordinate parallel work Bots start and supervise worker sessions; results and questions return to the bot chat. Source: https://kiro.dev/blog/crew-bots/ ### Schedules and events Scheduled jobs, webhooks and heartbeats trigger recurring work. The Gateway must be running. Source: https://kiro.dev/crew/ ### Custom dashboards Bots can present the state of their tasks and goals in customizable dashboards. Source: https://kiro.dev/crew/ ### Tool permissions and logs Crew combines approval modes, policies, a sandbox and an audit log. Their effect depends on the backend and configuration. Source: https://kiro.dev/docs/crew/security/ ## Example workflows ### Triage CI failures regularly - Your input: A repository, read access to CI and a clear escalation rule. - Possible result: A bot gathers findings and prepares scoped fixes. - What to check: Review the cause, diff, tests and approval before merging. ### Maintain recurring research - Your input: A topic, permitted sources and a schedule. - Possible result: A bot updates its briefing and flags pending decisions. - What to check: Check sources and freshness; correct outdated memories. ## Requirements and limits ### Continuous operation needs a host Requirement: A running service or container enables work while your laptop is closed. You manage hosting and resources. Source: https://kiro.dev/docs/crew/running-24-7/ ### Kiro access and credits Requirement: The default setup requires a signed-in Kiro CLI. Crew uses the Kiro subscription and its credits. Source: https://kiro.dev/docs/crew/ ### Alternative backends are in Preview Limitation: Backend selection requires Developer Mode. Capabilities differ; Pi, for example, receives no Crew MCP tools. Source: https://kiro.dev/docs/crew/features/agent-backends/ ### Approvals depend on the mode Limitation: Automatic and Autopilot do not request confirmation for each permitted call. Claude Code pre-approvals can bypass Crew checks. Source: https://kiro.dev/docs/crew/security/ ## A good fit if … - you want to delegate recurring development tasks to separate bots - you want to control memory and working state yourself - you want to combine schedules and events with inspectable work logs ## Less suitable if … - you want a fully managed service without your own Gateway - you expect Preview backends to have identical capabilities Editorial assessment based on provider documentation. Not a product test. ## Usable AI models Fixed choice across providers The standard Kiro CLI access offers Auto or a fixed choice of models from several providers. Bots and workers can use different models. Alternative harnesses are in Preview and are not rated here as generally available open model access. - Model families named: Anthropic (Claude), OpenAI (GPT), DeepSeek, Alibaba (Qwen), Moonshot (Kimi), MiniMax, Z.ai (GLM) - Model choice: Automatic or your choice - Your own API key: Not stated - Router or custom endpoint: Not stated - Local models: Not stated - Depends on: Plan, Admin approval, Region Documented · Sources accessed on: 2026-11-06 - [Kiro: Models](https://kiro.dev/docs/models/) - [Kiro Crew: Bots](https://kiro.dev/blog/crew-bots/) - [Kiro Crew: Agent backends (Preview)](https://kiro.dev/docs/crew/features/agent-backends/) ## Europe & regional availability Local operation and model processing are assessed separately. Reviewed scope: Your own host and Kiro model routing Reviewed on: 2026-11-06 ### Model routing depends on the account With requirements Your own EU host does not determine the model region. Individual subscriptions use the US endpoint; enterprise profiles in Frankfurt have model-dependent EU routes. According to the routing table, GPT-5.6 is processed globally through the US endpoint there too. Region covered: EU/EEA, United Kingdom and Switzerland Documented reason: Account type, profile region and selected model determine routing. Official source: https://kiro.dev/docs/models/ Official source: https://kiro.dev/docs/crew/ ## Accounts & permissions Based on provider documentation. Permissions depend on the account, integration and organization policies. Review the permissions actually requested before use. Configuration covered: Crew Gateway and standard Kiro access Sources reviewed: 2026-11-06 ### Account used Documented: Kiro CLI is signed in separately; the dashboard requires an access token. Evidence: https://kiro.dev/docs/crew/ Evidence: https://kiro.dev/docs/crew/running-24-7/ ### Admin approval Documented: Policies and profiles restrict permissions. Organizations can impose additional MCP controls. Evidence: https://kiro.dev/docs/crew/security/ ### Data access & team sharing Documented: Chat channels use authorized user identities. Tools operate within the approved access. Evidence: https://kiro.dev/docs/crew/security/ ## Prices and included usage Pricing checked: 2026-11-06 The software is open source. The standard model access uses Kiro credits; hosting costs are additional. ### Kiro Free USD 0 / month 50 monthly credits; limited model selection. Price source: https://kiro.dev/pricing/ ### Kiro Pro USD 20 / user / month 1,000 monthly credits; premium models. Price source: https://kiro.dev/pricing/ Further plans are available. Prices exclude applicable taxes; additional credits cost USD 0.04 each. Pricing details: https://kiro.dev/pricing/ ## Operation & oversight Based on public provider documentation. Assessments are editorial inferences from the stated properties. Actual time requirements and reliability have not been measured. Deployment covered: Your own Crew Gateway for ongoing work Sources reviewed: 2026-11-06 ### Files and working state Documented: Sessions and accepted background tasks survive Gateway restarts. Evidence: https://kiro.dev/docs/crew/system/ ### Sign-ins Documented: Kiro sign-in and channel credentials must remain valid; dashboard links have a limited lifetime. Evidence: https://kiro.dev/docs/crew/running-24-7/ ### Starting further runs Documented: Schedules, webhooks and heartbeats can trigger further tasks. Evidence: https://kiro.dev/crew/ ### Human intervention Documented: Bots ask questions in the ongoing chat; independent work can continue. Evidence: https://kiro.dev/blog/crew-bots/ ### Detecting errors Documented: The System view and Crew Log show task status, tools and approvals. Evidence: https://kiro.dev/docs/crew/system/ ### Resuming after failures Documented: The durable queue resumes accepted work after a Gateway restart. Evidence: https://kiro.dev/docs/crew/system/ ### Editorial assessment Suitable for supervised ongoing tasks with clear permissions; hosting, costs and results need regular review. ## Memory and conversation continuity Scope: Crew memory and private bot memories Sources accessed: 2026-11-06 Persistent memories, separate bots and visible controls; session modes govern reading and writing. ### Storage and scope Documented: Crew stores preferences and project context as Markdown and semantic and episodic memories in SQLite on the Gateway host. - [Kiro Crew: Memory](https://kiro.dev/docs/crew/features/memory/) ### Selection and timing Documented: Explicit lessons are saved immediately; consolidation runs after 30 messages or three hours of inactivity. - [Kiro Crew: Memory](https://kiro.dev/docs/crew/features/memory/) ### Inspect, edit and delete Documented: Bot memories can be inspected, edited and deleted. Incognito reads without writing back; Temporary neither reads nor writes memory. - [Kiro Crew: Bots](https://kiro.dev/blog/crew-bots/) - [Kiro Crew: Sessions](https://kiro.dev/docs/crew/chat/sessions/) ### Agent and channel boundaries Documented: Bots have private memory. Regular channel sessions share long-term knowledge but keep separate conversation context. - [Kiro Crew: Bots](https://kiro.dev/blog/crew-bots/) - [Kiro Crew: Sessions](https://kiro.dev/docs/crew/chat/sessions/) ### Retrieval Documented: Preferences are injected at session start. Semantic search combines vectors and keywords; keyword search remains available without embeddings. - [Kiro Crew: Memory](https://kiro.dev/docs/crew/features/memory/) ### Session reset and compaction Partly documented: Sessions survive restarts. The work ledger preserves the goal, next step and artifact references across compaction. The exact deletion boundaries of a full reset remain unspecified in the reviewed sources. - [Kiro Crew: Sessions](https://kiro.dev/docs/crew/chat/sessions/) - [Kiro Crew: System & storage](https://kiro.dev/docs/crew/system/) ## Context Background work requires an available Gateway and host. Model access, permissions and costs must fit the task. Official product documentation reviewed. Workflows and assessments are editorial; hands-on product tests are outside the scope of this directory. ## Sources - https://kiro.dev/crew/ - https://kiro.dev/blog/crew-bots/ - https://kiro.dev/docs/crew/ - https://kiro.dev/docs/crew/features/memory/ - https://kiro.dev/docs/crew/chat/sessions/ - https://kiro.dev/docs/crew/security/ - https://kiro.dev/docs/crew/system/ - https://kiro.dev/docs/crew/running-24-7/ - https://kiro.dev/docs/crew/features/agent-backends/ - https://kiro.dev/docs/models/ - https://kiro.dev/pricing/ - https://github.com/kirodotdev/KiroCrew ## Provider Kiro Crew is developed in the official Kiro project under Apache-2.0. - Operator: Amazon Web Services (AWS) - Organization: Provider and open-source community - Website: https://kiro.dev/crew/ - Reviewed: 2026-11-06 - Repository: https://github.com/kirodotdev/KiroCrew - GitHub-Stars: 4809 (2026-11-06) - License: Apache-2.0 - Provider source: https://kiro.dev/crew/ - Provider source: https://github.com/kirodotdev/KiroCrew [Sources and method](https://ki-agentenkompass.de/en/sources/)