kagent

Open source toolUnknown

Bring AI agents to every cluster you run

Homepage Docs Repository

Scanned 8 Oct 2026 · Schema v5 · features v21

Licence: Apache-2.0
Self-hosted; you pay for the infrastructure.
Free tier: Free to run
Isolation: Kata / gVisor
hardened or user-space kernel
Regions: —
Not disclosed on the public surface.
Capabilities: 21 / 61
13 partial · 11 fail · 0 not assessed
Agent readiness: 8/10 · 7/9
read · drive, pre-login surface

Advertised for

Kubernetes operations · Multi agent workflows · Coding agents · Standalone sandboxes

Built for

Developers, then technical

· Explicitly 'Built specifically for DevOps and platform engineers' — kubectl / Helm / CRD / GitOps-first onboarding
· Deep cloud-native vocabulary: pods, RBAC, CRDs, Istio/Ambient Mesh, gVisor, WorkerPool/Actor, OpenTelemetry, ArgoCD
· Install is CLI + Helm into a Kubernetes cluster (kind for local); agents defined as YAML applied with kubectl

01 Runtime Each fact carries its source; dots show confidence. Open a row for the evidence.

Isolation & lifecycle

Isolation primitive

Kata / gVisorhardened or user-space kernel

Persistence

Postgres holds platform state and optional pgvector memory; idle Actors persist memory/filesystem snapshots in object storage.

Snapshots / forks

Full/Data snapshots plus documented Session checkpoint and fork operations.

Resources & limits

Storage

PostgreSQL, optional pgvector agent memory, and operator-provided snapshot object storage.

Limits & quotas

Operator WorkerPool capacity; standalone sandboxes default to 1 CPU/1Gi, TTL 1h and maxTTL 24h, configurable at installation.

GPU

Access

Deploy interface

helmclidashboardapi

SDK languages

GoPython

Network & secrets

Egress controls

Default-deny per-Actor gateway policy derived from model/tool/skill/telemetry destinations; TLS termination enforces hostname rules.

Secrets

Kubernetes Secrets stay outside agent environments; the egress gateway injects ModelConfig API keys and RemoteMCPServer secret-backed headers.

Bring your own cloud

Placement & use

Jurisdiction

Advertised workloads

Kubernetes operationsMulti agent workflowsCoding agentsStandalone sandboxes

02 Capabilities 21 pass · 13 partial · 11 fail · 61 of 61 assessed

Supported — Documented on the public surface and verified.
Partially supported — Exists with caveats: gated, thin, or enterprise-only.
Not supported — Checked and not offered.
Not assessed — Scanned before this capability was tracked. Not a lack.

Compute & storage 4/8

Persistent diskStorage that survives restart/stop (not ephemeral-only).

Partially supported

Object storageManaged S3-compatible / blob object store the platform PROVIDES to a user's workloads (buckets, presigned URLs). DISTINCT from persistent-disk (block/volume attached to an instance) and from a self-host dependency (source.self_host.deps[]=["object-storage"] is an infra dep to RUN an OSS tool — the opposite direction). SCOPE varies and matters: native first-class (Railway Buckets, Cloudflare R2) vs a sibling SKU / extension (Fly Tigris, Bunny Storage, DigitalOcean Spaces) — record which in the evidence quote / note. Admitted v15.

Not applicable

Snapshots / forkingSnapshot or copy-on-write fork of a running instance (filesystem-only counts). Whether the snapshot also restores memory + running processes is recorded separately by memory-snapshot (v19).

Supported

Live-state (memory) snapshotsA snapshot / checkpoint / pause captures a running instance's MEMORY and PROCESS state and restores it, so running processes continue where they left off instead of a freshly booted instance with the same files. Counts whether user-triggered (snapshot, checkpoint, pause-to-storage) or used to restore pre-initialised state (memory snapshots for fast cold starts). Distinct from snapshots-forks (any snapshot or fork, including filesystem-only; a product passing that key may still fail this one) and from scale-to-zero (an idle-cost property; suspension may or may not keep memory). Actor multiplexing (see scale-to-zero) records its live-state half here — only when the suspended actor's MEMORY/process state is what comes back, not a freshly booted worker with the actor's files.

Supported

Scale to zeroIdle instances suspend at no / negligible cost. ACTOR MULTIPLEXING (many suspended stateful actors sharing a small pool of pre-warmed workers, woken on request — Agent Substrate, AX, kagent; Rivet's hibernating actors) has NO key of its own (proposal REJECTED v20) — score the idle-cost half HERE and the live-state half under memory-snapshot; the oversubscription ratio / density is a performance characteristic the registry does not key.

Supported

GPU computeGPU-backed instances available.

Not supported

Multi-regionCompute placeable in more than one geographic region.

Partially supported

Sandbox templates / custom imagesBuild reusable sandbox/VM images from an OCI image or Dockerfile (a base you fork instances from).

Supported

Cloud-init / ignition configCustomise a VM / microVM / instance at boot via cloud-init (or ignition) user-data — the same boot-metadata interface as AWS/Azure/GCP. Inject packages, users, SSH keys, files, and run-commands declaratively at launch. Distinct from sandbox-templates (a PREBUILT image you fork instances from) — this is PER-LAUNCH declarative customisation, not a baked image.

Not applicable

Desktop / computer-use sandboxA virtual desktop (GUI) for LLM computer-use — browser + desktop automation, often via noVNC/VNC.

Not applicable

Selectable isolation runtimeThe user chooses the isolation runtime per workload / environment / sandbox from more than one option (e.g. microVM vs gVisor vs plain container, native namespaces vs gVisor, gVisor vs VM). Records the CHOICE; the isolation_primitive facet still records the best-evidenced primary. A platform that switches runtime automatically without a user choice (e.g. gVisor forced for GPU) does not pass on that alone.

Not supported

Networking & delivery 2/4

Automatic HTTPSTLS endpoints provisioned automatically per workload.

Not applicable

Built-in reverse proxyPlatform routes external traffic to workload services.

Partially supported

Custom domainsBring-your-own domain via CNAME/DNS.

Not applicable

Public preview URLsShareable per-deploy / per-branch URLs.

Not applicable

Direct public IPv6Per-instance stable public IPv6 for raw TCP/UDP (beyond a managed HTTPS proxy) — e.g. mosh / raw sockets.

Not applicable

Egress controlsConfigurable OUTBOUND network policy — egress firewall / allowlist / egress proxy on traffic LEAVING a workload. Scoped to OUTBOUND only (v15): private-network REACHABILITY between a user's own services is private-networking, not this. Static-egress-IP belongs here (an outbound concept).

Supported

Private networkingA platform-internal private network between a user's OWN services — service-to-service connectivity, VPC, service mesh, and/or service discovery, not exposed to the public internet. DISTINCT from egress-controls (OUTBOUND firewall/allowlist), reverse-proxy (PUBLIC inbound ingress), and direct-ipv6 (per-instance PUBLIC addressing). Admitted v15.

Supported

Email send / receiveSend and/or receive email from the workload.

Not supported

Edge KV storeA globally-replicated key-value store read from WITHIN an edge runtime at the PoP (low-latency, eventually-consistent) — Cloudflare Workers KV, Deno KV, Fastly KV Store, Akamai EdgeKV, Vercel Edge Config. Distinct from first-party-db (a managed DATABASE service you provision and connect to) and from http-query-api (HTTP access to an external store) — edge-kv is colocated with the edge compute and read at the edge.

Not applicable

Developer access 4/9

Public APIDocumented programmatic API.

Supported

OpenAPI specMachine-readable OpenAPI / Swagger spec published.

Partially supported

CLIOfficial command-line tool.

Supported

SDKOfficial client library/libraries.

Partially supported

SSH accessDirect SSH into workloads.

Not supported

Filesystem APIRemote file operations over the API (list/read/write/watch/copy/upload) without SSH.

Supported

Git deployDeploy from a Git repo / push-to-deploy.

Supported

Private container registryHosted private image registry.

Not supported

E2B-compatible APIDrop-in compatibility with the E2B sandbox API/SDKs — existing E2B client code runs against this product by changing the endpoint/domain + key, with no business-code changes. E2B itself passes as the reference implementation. Distinct from public-api / sdk (the product has its own API — this key says it also speaks another vendor's) and from self-host-oss (a self-hosted E2B-compatible server is scored on both).

Not supported

HTTP data APIQuery / read / write the stored DATA over plain HTTP/REST (or GraphQL) WITHOUT the native protocol driver (Postgres wire / Redis RESP) — agent- and edge-friendly data access via fetch(). The DATA plane; distinct from public-api (a control-plane / management API). Auto-generated-from-schema REST/GraphQL (PostgREST, pg_graphql) is a flavour of this.

Not applicable

AI & agents 7/12

Built-in coding agentShips a first-party AI coding agent in the product.

Partially supported

Agent skillsInstallable / declarable agent skills or tools.

Supported

AI app builderPrompt-to-app / vibe-coding generation surface.

Not supported

MCP serverExposes a Model Context Protocol server for agents.

Supported

MCP clientActs as an MCP client / consumes MCP tools.

Supported

MCP-server hostingRun/host MCP servers as a workload on the platform (distinct from exposing a first-party MCP control server — that is mcp-server).

Supported

LLM gatewayRoutes LLM API calls through the platform (multi-provider).

Partially supported

BYOK for LLMsBring-your-own LLM provider keys.

Supported

Model inference API (first-party)A FIRST-PARTY hosted LLM/model inference API — the provider serves its OWN catalogue of (usually open-weight) models behind a (typically OpenAI-compatible) endpoint, billed per token. Distinct from llm-gateway (a multi-PROVIDER proxy/router for YOUR calls to external providers like OpenAI/Anthropic) and from byok-llm (bring your own provider keys) — here the platform IS the inference provider, serving its own models.

Not supported

Autonomous QA agentA first-party AUTONOMOUS AI agent that QA-TESTS a running app / implementation — exercising it like a real user (clicking through flows, checking behaviour) and reporting bugs/defects (often back to a coding agent, or before a human review). Distinct from built-in-agent (a code-AUTHORING agent — this one VERIFIES the result, it does not write the feature; a product can ship both), from ai-guided-code-review (organises a diff for a HUMAN to review — not an autonomous tester), and from observability (runtime metrics/logs, not active testing).

Not supported

Credential-injection proxyThe platform holds real secrets OUTSIDE the workload and injects them on the wire (an egress / broker proxy swaps a placeholder or phantom token for the real credential on matching outbound requests), so code running in the sandbox — e.g. an untrusted agent — never sees or can exfiltrate the real key. Distinct from secrets-management (storage + injection INTO the workload as env vars/files — the workload holds the secret) and from egress-controls (outbound allow/deny POLICY; a product can have an allowlist without injecting credentials, and vice versa).

Supported

Preinstalled coding agentThe platform installs and launches a THIRD-PARTY coding agent (a harnesses.yml entry — Claude Code, Codex, OpenCode, Pi ...) inside a workspace/sandbox it provisions, with the agent's auth wired in (your own account or API key), so you start an agent session rather than an empty box. Distinct from built-in-agent (a FIRST-PARTY agent the vendor ships) and from harnesses[] (which records that a harness CAN target the product, not that the product runs it for you).

Supported

Data 1/1

First-party databasePlatform-provided managed database.

Not applicable

Database branchingCopy-on-write branch of a DATABASE (schema + data) as a dev-lifecycle primitive — branch your data like code, usually instant via a metadata operation. Distinct from snapshots-forks (forks a running compute INSTANCE / VM state) and from preview-environments (a full per-PR environment that may USE a DB branch). For a data product a CoW branch maps HERE, not to snapshots-forks.

Not applicable

Vector searchFirst-class storage, indexing, and similarity search of vector embeddings — a native vector type/index or a managed vector DB. The data-layer primitive for RAG / semantic retrieval.

Supported

Ephemeral instant DB (no-signup)Provision a working, usable instance via a single command / HTTP call with NO account, signup, or card — typically time-boxed (expires unless claimed). A zero-onboarding path an agent can hit directly.

Not applicable

Platform & operations 1/4

Scheduled tasksCron / scheduled job primitive.

Not supported

Durable workflowsDurable execution / long-running workflow primitive.

Partially supported

Preview environmentsA full EPHEMERAL environment spun up per PR / branch — compute + addons + (often) a DB branch — torn down on merge. More than preview-urls (which is just a shareable URL for one deploy).

Not applicable

Built-in observabilityFirst-party metrics / logs / tracing / APM / profiling included in the platform (no third-party stack required to see what a workload is doing).

Supported

Managed CI/CD runnersPlatform-hosted, drop-in CI/CD runner VMs (GitHub Actions / GitLab), typically faster and cheaper than the CI vendor's own hosted runners, billed per runner-minute; isolated + ephemeral per job. Distinct from git-deploy (push-to-deploy of an APP) and scheduled-tasks (cron) — this is managed CI COMPUTE you point your existing pipelines at.

Not applicable

Project template marketplaceA marketplace / gallery of prebuilt PROJECT STARTER templates (apps or websites) a user clones or launches a new project from and then customises — the "start from a template" surface of a builder / PaaS. Distinct from sandbox-templates (a COMPUTE base IMAGE you fork runtime INSTANCES from — Dockerfile / OCI), from app-builder (the prompt-to-app GENERATION surface — a product can have both), and from one-click OSS-app catalogues / helm-addon-marketplace (deploy a FINISHED app as-is, not a starter you build on).

Partially supported

Organisation & security 2/7

Built-in authPlatform-provided authentication / IAM for workloads.

Not supported

SSO / SAMLEnterprise single sign-on.

Partially supported

Secrets managementFirst-class secrets storage / injection.

Partially supported

TeamsMulti-user team accounts.

Partially supported

SharingShare a running workload/resource via link (optionally gated).

Supported

BYOCRuns inside the customer's own cloud account.

Not applicable

Deploy into customers' VPCsA SaaS vendor uses the platform to deploy and operate ITS software inside ITS CUSTOMERS' cloud accounts / VPCs (nested or "customer BYOC"), managed from the vendor's control plane. Distinct from byoc (your own workloads in your own cloud account).

Not applicable

Workload identityThe platform gives each workload its own verifiable identity (OIDC token, SPIFFE SVID, cloud IAM federation) that it exchanges for short-lived credentials, so no long-lived static secret is stored or injected. Distinct from secrets-management (stores / injects static secrets) and credential-injection-proxy (swaps real secrets in on the wire).

Partially supported

Self-host / OSS coreOpen-source core that can be self-hosted.

Supported

Proposed, not verified

Capabilities the scan saw but the registry does not track yet. A key is admitted only once two distinct products exhibit it. These are not scored.

Agents-as-CRDs — declarative AI agents as first-class Kubernetes custom resources (kubectl / GitOps / RBAC lifecycle)agents-as-crds
“Define agents as Kubernetes CRDs — versioned in Git, reviewed in PRs, rolled out with the same tools you already use. kagent makes agents another first-class workload, not a new platform to learn.” kagent.dev
Agent-to-Agent (A2A) — agents discover and invoke each other to compose multi-agent workflows with first-class delegationagent-to-agent
“Agent-to-Agent (A2A) — Agents discover and invoke each other. Compose multi-agent workflows with first-class delegation.” kagent.dev
Human-in-the-loop — tool-approval gates, agent-initiated questions, and cascading HITL so humans stay in controlhuman-in-the-loop
“Human-in-the-loop — Tool approval gates, agent-initiated questions, and cascading HITL — humans stay in control.” kagent.dev
Context compaction — automatic summarization of long histories to stay within token budgetscontext-compaction
“Context compaction — Auto-summarization of long histories. Agents stay coherent in extended conversations without blowing token budgets.” kagent.dev

03 Agent readiness Pre-login surface · counts, not a grade

Can an agent read it?

parse-ability · docs, metadata, machine-readable surfaces

8/10

5 full · 3 partial · 80% of applicable checks

llms.txt
Supported
“/docs/llms.txt and /docs/kagent/1.x/llms.txt return substantive indexes linking versioned guides, APIs and per-page .md sources. Apex /llms.txt remains 404.”
llms-full.txt
Not supported
“/llms-full.txt returns 404; indexes link individual markdown pages, not an advertised full-document corpus.”
robots.txt AI policy
Partially supported
“robots.txt (200) uses the Cloudflare 'content-signals' framework (search/ai-input/ai-train) with no explicit yes/no on the AI signals = 'neither grants nor restricts' = default-allow; no AI-UA Disallow.”
JSON-LD
Not supported
“No <script type=application/ld+json> block on the homepage.”
Sitemap
Supported
“/sitemap.xml returns HTTP 200 with 23 loc entries.”
Meta tags
Supported
“Descriptive <title> + full OG (title/description/image 2400x1260) + twitter card.”
Markdown negotiation
Partially supported
“Versioned .md pages return clean text/markdown; partial credit for explicit URL variants, not a verified Accept-header negotiation mechanism.”
Content efficiency
Partially supported
“Homepage HTML 208357 bytes; visible text (HTMLParser excluding script/style) 8652 UTF-8 bytes = 4.15%, rubric 1–5% band. Inline SVG/Next.js payload inflate HTML.”
Docs discoverable
Supported
“Docs first-class in the header nav ('Docs' dropdown) + footer; DocSearch (Algolia) search present; reachable without JS via sitemap.”
Machine-readable pricing
Not applicable
“No pricing surface — kagent is free Apache-2.0 OSS (self-host). The /enterprise page points to Solo.io's separate commercial distribution, not kagent pricing tiers. Dropped from the denominator.”
No anti-bot wall
Supported
“Homepage, current versioned docs, robots and sitemap fetched without a challenge on 2026-10-08. Old unversioned docs return 500; recorded as a URL-migration issue, not a bot block.”

Can an agent drive it?

act-ability · APIs, SDKs, MCP, credentials, webhooks

7/9

4 full · 3 partial · 78% of applicable checks

Public API
Supported
“Documented controller gRPC, Kubernetes v1alpha3 resources, A2A and MCP; self-hosted endpoints, not a hosted service.”
OpenAPI spec
Partially supported
“CRD reference has field and validation schemas; no aggregate downloadable OpenAPI found in that reference or common paths.”
MCP server
Supported
“Documented controller Streamable HTTP /mcp on port 8083; concrete Claude Code/Cursor configs and curl examples for Session/checkpoint/sandbox tools.”
SDK packages
Not supported
“Go/Python agent engines do not by themselves verify official platform-client SDK packages. Prior full credit for runtime language count was unsupported.”
Auth documented
Supported
“substrate-runtime/identity.md explicitly documents insecure and trusted-proxy modes, lack of signature/expiry verification and permissive authorization. Documentation credit is NOT security approval.”
Self-serve credentials
Not applicable
“Self-hosted OSS — there is no hosted signup / credential-issuance surface; you install it and supply your own LLM keys. Dropped from the denominator (needs a hosted endpoint).”
Webhooks
Not supported
“Messaging-channel event integration (Slack/Discord/Telegram) exists but no documented webhook endpoint incl. signing was found.”
CLI tool
Supported
“Official kagent binary and command reference; sandbox guide demonstrates install and lifecycle/file/process commands.”
Error model
Partially supported
“MCP examples document JSON-RPC Invalid params (-32602), mismatched elicitation keys and polling transitions; no complete platform-wide error schema verified.”
Rate limits documented
Not applicable
“No vendor-hosted API or vendor rate-limit headers; self-hosted installation capacity and sandbox TTL limits are documented separately.”
Agent config files
Partially supported
“Documented skills and agent configuration. No additional AGENTS.md or well-known-agent surface verified in this capture.”

04 Coding harnesses Which agents and IDEs are documented to run here

Advertised integrations 3

Codex CLI OpenAI
Supported
“1.x Harness codex runtime runs the Codex coding agent.”
Source kagent.dev
Claude Code Anthropic
Supported
“1.x Harness claude runtime runs the Claude coding agent.”
Source kagent.dev
Cursor Anysphere
Supported
“Documented Cursor MCP settings connect to the controller /mcp endpoint; client integration, not a Cursor runtime backend.”
Source kagent.dev

Plausible compatibility 0

Derived, not stored: the product passes SSH access, so any CLI coding agent is installable inside it. This is plausible compatibility, not a supported integration.

No general shell access documented, so nothing is derived. Absence is not unsupported.

05 Compliance & regions Standards held, with ordinal rigor; never a score

Certifications held 0

No attestations found on the public surface. Empty means nothing is advertised, after checking, not that nothing was checked.

Regions —

No region list captured.

Legal

06 Open source Apache-2.0 · fetched by machine, not quoted from marketing

Licence & repository

Licence: Apache-2.0OSI
Primary language: Go
Stars: 3,965
Forks: 825
Created: 21 Jan 2025
Archived: No

Activity

Last commit: 8 Oct 2026
Latest release: v1.0.0-alpha9 · 7 Oct 2026
Issues open / closed: 109 / 819 · 88% closed

v1.0.0-alpha9 on 2026-10-07; alpha8 Oct 5, alpha7 Oct 1. GitHub marks these isPrerelease=false, but the tag names are alpha. Latest non-alpha tag among eight fetched releases: v0.10.3 (2026-09-28). The 1.x version table still pins alpha8.

Fetched via gh on 8 Oct 2026.

Self-hosting

Methods: helmkubernetesbinary
You operate: kubernetespostgres

Self-host only: Helm 3, kubectl, jq, openssl and kubectl-ate; install Agent Substrate and create CA/JWT identity material before kagent. Bring model credentials, PostgreSQL (pgvector for agent memory), object storage and a gVisor WorkerPool. Default install is evaluation-oriented, not production-ready. Installation says Kubernetes 1.37+ with certificates.k8s.io/v1beta1, while version support says CI tests 1.35: unresolved docs inconsistency, not a verified deployment requirement. 0.10.x requires a fresh cluster/database for 1.x; no in-place migration or conversation-history carry-over. Cost is Apache-2.0 software plus operator infrastructure and model usage.

Documentation signals 7/8

SupportedQuickstart
SupportedAPI reference
SupportedSelf-host guide
SupportedChangelog
Not supportedOpenAPI
Supportedllms.txt
SupportedExamples
SupportedArchitecture doc

07 Company & presence Facts only, not scored

Company

Headquarters: Not disclosed
Founded: 2025 (kagent created at Solo.io; CNCF Sandbox)
Team size: Community project — maintained by Solo.io + contributors from Google, Confluent, Adobe, Red Hat, IBM, Microsoft
Hiring: n/a — open-source CNCF project (contributions via GitHub/Discord, not a jobs board)

Notable people

Solo.io (creator; 'built by the founders of Istio')

Activity

Last release: v1.0.0-alpha9 (2026-10-07) github.com
Last blog post: Blog available; latest publication date not established kagent.dev

Presence 6

Links found on the public surface scanned; not an exhaustive search.

08 Sources & method How this entry was produced

Public surface only

Everything here was read from kagent.dev's unauthenticated pages, docs, pricing and repository on 8 Oct 2026. Nothing behind a login was used.

Statuses, not scores

Capabilities are pass, partial or fail against a written rubric (features v21). “Not assessed” means the scan predates the key.

Confidence is separate

Each fact carries its own confidence. Low-confidence facts are shown, never hidden, so you can weigh them.

Schema: scan v5
Features registry: v21
Keys tracked today: 61
Assessed on this entry: 61

Something wrong on this page?

Nearby in the atlas

All Unknown products →