Recurse

HostedServerless

A serverless harness for agents to create custom agents and deploy them as tools, MCPs, or bots.

Homepage Docs Pricing Repository

Scanned 26 Sep 2026 · Schema v5 · features v19

From: Free
Prepaid account credit, pay per request: LLM tokens at provider rates (no markup) + serverless execution per minute. No subscription or idle charges. New accounts get $5 credit, no card required. Top-ups $5–$500.
Free tier: Yes
No card required.
Isolation: Not disclosed
the public docs do not say
Regions: —
No region information on the public surface; execution runs on an unnamed third-party serverless provider.
Capabilities: 6 / 57
4 partial · 41 fail · 4 not assessed
Agent readiness: 10/11 · 8/11
read · drive, pre-login surface

Advertised for

Custom LLM agents that iterate against verifiers (search/optimisation loops) · ML model tuning and predictive modelling · Design space search (CAD/brackets, RNA sequence design, 3D object reconstruction) · Game level generation with solver/simulator checks · Agents deployed as MCP tools for coding agents (Codex, Claude Code)

Built for

Developers, then technical

· CLI + Python 3.14 + uv toolchain; agent.yaml manifests
· hero shows YAML and terminal output
· usage pricing per token and per minute

01 Pricing Prepaid account credit, pay per request: LLM tokens at provider rates (no markup) + serverless execution per minute. No subscription or idle charges. New accounts get $5 credit, no card required. Top-ups $5–$500.

Tier	EntryPay per request
Price	Execution $0.013/min; GPT-5.6 Luna $0.20/$1.20, Terra $2/$12, Sol $4/$20; GPT-6 Astra $10/$50, Luna $0.10/$0.50, Sol $2/$10 per 1M input/output tokens
Includes	$5 starter credit for new accounts. Non-preemptible direct runs cost 3× for Function CPU and memory.

per 1M input/output tokens (model calls) + per minute (serverless execution)·Card required: No·recurse.run

Overages: Pure metering against prepaid credit; when the balance can't fund active work, runs are cancelled with account_balance_exhausted.

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

Isolation & lifecycle

Isolation primitive

Not disclosedthe public docs do not say

Persistence

Ephemeral per run: each run or deployed tool call is a fresh serverless run; results and artifacts are retained 24 h, then downloads are denied (metadata stays visible).

Snapshots / forks

None documented

Resources & limits

Storage

Run workspace (recurse.context().workspace) → downloadable artifacts with size + SHA-256, retained 24 h. Source archive ≤64 MiB, ≤1,024 members.

Limits & quotas

Per run: 0.125–16 CPU (default 1), 512–16384 MiB (default 1024); up to 24 h per provider attempt (3 min reserved for finalisation), no automatic restart; direct runs preemptible unless --non-preemptible (3× Function CPU/memory price); artifacts 24 h; source archive 64 MiB / 1,024 files; secret ≤32 KiB. Runs stop when account balance is exhausted.

GPU

Access

Deploy interface

clisdkmcp

SDK languages

Python

Network & secrets

Egress controls

Secrets

Encrypted account runtime secrets (`recurse secret set`, stdin input, no read-back), bound per run/deployment as env vars (`--secret ENV=NAME`); supplied to tool execution only, not to model calls or packaging; max 32 KiB per value.

Bring your own cloud

Placement & use

Jurisdiction

unknown

Advertised workloads

Custom LLM agents that iterate against verifiers (search/optimisation loops)ML model tuning and predictive modellingDesign space search (CAD/brackets, RNA sequence design, 3D object reconstruction)Game level generation with solver/simulator checksAgents deployed as MCP tools for coding agents (Codex, Claude Code)

03 Capabilities 6 pass · 4 partial · 41 fail · 57 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 1/9

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

Not 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 supported

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).

Not 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.

Not 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.

Not supported

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

Not 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 supported

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 assessed

Networking & delivery 0/8

Automatic HTTPSTLS endpoints provisioned automatically per workload.

Not supported

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

Not supported

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

Not supported

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

Not supported

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

Not supported

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).

Not 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.

Not 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 2/9

Public APIDocumented programmatic API.

Partially supported

OpenAPI specMachine-readable OpenAPI / Swagger spec published.

Not supported

CLIOfficial command-line tool.

Supported

SDKOfficial client library/libraries.

Supported

SSH accessDirect SSH into workloads.

Not supported

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

Not supported

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

Not 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 2/11

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.

Not supported

MCP clientActs as an MCP client / consumes MCP tools.

Not 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.

Not 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).

Not 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).

Not assessed

Data 0/1

First-party databasePlatform-provided managed database.

Not supported

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.

Not applicable

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 0/6

Scheduled tasksCron / scheduled job primitive.

Not supported

Durable workflowsDurable execution / long-running workflow primitive.

Not 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 supported

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).

Partially 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 supported

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).

Not supported

Organisation & security 1/7

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

Not supported

SSO / SAMLEnterprise single sign-on.

Not supported

Secrets managementFirst-class secrets storage / injection.

Supported

TeamsMulti-user team accounts.

Not supported

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

Not supported

BYOCRuns inside the customer's own cloud account.

Not supported

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 assessed

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).

Not assessed

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

Not 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.

Verifier-gated agent optimisation loop (agent evaluates candidates against external verifiers and keeps the best feasible result)verifier-loop-harness
“Separate feasibility from quality. All checks must pass before a candidate can win on its objective … Preserve the best feasible candidate when later attempts regress.” recurse.run
Deploy an agent as an MCP tool with a schema-typed input/output contractagent-as-tool-deploy
“The manifest becomes the MCP tool's interface. `metadata.name` supplies its name … `inputs` defines what callers supply, and `outputs` defines the structured result.” recurse.run

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

Can an agent read it?

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

10/11

8 full · 2 partial · 91% of applicable checks

llms.txt
Supported
“/llms.txt (5.7 KB) links SKILL.md, every docs page, pricing, examples, use cases, blog and the agent.yaml schema.”
llms-full.txt
Supported
“/llms-full.txt (208 KB) = SKILL.md + all public pages (artifacts/llms-full.txt.POINTER.md).”
robots.txt AI policy
Partially supported
“robots.txt: 'User-Agent: * / Allow: / / Allow: /llms.txt / Allow: /SKILL.md' — default-allow with agent files called out, but no AI user agents named.”
JSON-LD
Not supported
“No <script type="application/ld+json"> in the homepage HTML.”
Sitemap
Supported
“/sitemap.xml: 26 URLs (docs, pricing, use cases, blog).”
Meta tags
Supported
“Descriptive <title>, meta description, full OpenGraph + Twitter card with 1200×630 image.”
Markdown negotiation
Partially supported
“Accept: text/markdown returns HTML and <page>.md 404s, but every page has a markdown twin at /llms/<page>.md (linked from llms.txt).”
Content efficiency
Supported
“Homepage ~9.1 KB readable text / 74 KB HTML ≈ 12% (server-rendered Next.js).”
Docs discoverable
Supported
“'Docs' is the first item in the primary header nav (plain <a href="/docs">).”
Machine-readable pricing
Supported
“/pricing (and /llms/pricing.md) is a per-model token table plus per-minute execution rate.”
No anti-bot wall
Supported
“Behind Cloudflare, but curl and firecrawl fetched every public page with no challenge.”

Can an agent drive it?

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

8/11

2 full · 6 partial · 73% of applicable checks

Public API
Partially supported
“Python API reference in the SDK repo (docs/reference/api.md) and a documented CLI; the underlying HTTP service API is mentioned ('The CLI and API…') but not documented.”
OpenAPI spec
Not supported
“/openapi.json 404; only a JSON Schema for agent.yaml (/schema/agent.yaml.json).”
MCP server
Not supported
“No first-party MCP server for operating Recurse. `recurse mcp serve` exposes a customer's deployed agent (hosting, not a control server); /.well-known/mcp 404.”
SDK packages
Partially supported
“One official SDK: Python (recurse-sdk on PyPI).”
Auth documented
Partially supported
“Documented CLI browser login (GitHub/Google) storing a device credential in the OS keychain, plus runtime-secret docs; no API keys or scoped tokens.”
Self-serve credentials
Partially supported
“`recurse login` needs a human browser sign-in; the coding agent can then act. No programmatic key issuance. Deployment access is private beta ('deployment access is currently limited').”
Webhooks
Not supported
“No webhooks documented.”
CLI tool
Supported
“Official `recurse` CLI (login, run, deploy, status, secret, billing, mcp serve).”
Error model
Partially supported
“Run failures carry a stable error.type / error.code with a safe message in a versioned YAML snapshot (e.g. engine/infrastructure_failed, account_balance_exhausted, secret_unavailable) — a CLI-level model, not a JSON HTTP error model.”
Rate limits documented
Partially supported
“Resource and duration limits documented (Costs and limits); no request rate limits or headers.”
Agent config files
Supported
“Several surfaces: /SKILL.md (coding-agent skill), Claude Code/Codex/Cursor plugin manifests (recurse-plugins), an AGENTS.md/CLAUDE.md trigger snippet on /install, llms.txt.”

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

Advertised integrations 5

Claude Code Anthropic
Supported
“Claude Code supports project skills in `.claude/skills/<name>/SKILL.md` … curl -fsSL https://recurse.run/SKILL.md -o .claude/skills/recurse/SKILL.md; `claude mcp add recurse -- recurse mcp serve <deployment-id>`; Claude Code plugin in synnada-ai/recurse-plugins.”
Source recurse.run
Codex CLI OpenAI
Supported
“For Codex, set `startup_timeout_sec = 180` and `tool_timeout_sec = 87000`; `codex mcp add recurse -- recurse mcp serve <deployment-id>`; `codex plugin add recurse@synnada-recurse`.”
Source recurse.run
Cursor Anysphere
Supported
“One Recurse skill for Cursor, Codex, and Claude Code … Copy `plugins/recurse` into `~/.cursor/plugins/local/recurse`, then restart Cursor.”
Source github.com
Hermes Agent Nous Research
Supported
“'Give this to your agent' lists Codex, Claude Code, Hermes and OpenClaw with `set up https://recurse.run/SKILL.md`.”
Source recurse.run
Low-medium confidence: a logo on the landing page, no Hermes-specific guide.
OpenClaw Peter Steinberger
Supported
“'Give this to your agent' lists Codex, Claude Code, Hermes and OpenClaw with `set up https://recurse.run/SKILL.md`.”
Source recurse.run
Low-medium confidence: a logo on the landing page, no OpenClaw-specific guide.

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.

06 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 —

Regions are not disclosed on the public surface.

No region information on the public surface; execution runs on an unnamed third-party serverless provider.

Legal

Privacy notice Terms of service

07 Company & presence Facts only, not scored

Company

Headquarters: unknown
Team size: unknown
Hiring: not hiring publicly

Notable people

Mehmet Ozan Kabak — Co-founder & CEO (github.com/ozankabak)
Sami Can Tandogdu — Co-founder & COO
Team includes core contributors to Apache DataFusion (synnada.ai)

Activity

Last release: recurse-sdk 0.2.3 on PyPI 2026-09-24 (0.1.7 → 0.2.3 in eight days) pypi.org
Last blog post: 2026-08-20 (three posts, all that date) recurse.run

Presence 7

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 recurse.run's unauthenticated pages, docs, pricing and repository on 26 Sep 2026. Nothing behind a login was used.

Statuses, not scores

Capabilities are pass, partial or fail against a written rubric (features v19). “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: v19
Keys tracked today: 61
Assessed on this entry: 57

Something wrong on this page?

Nearby in the atlas

All Serverless products →