REA Environments on AgentEnv
50 replay environments REA built from app sessions
50 RL environments REA built from recorded browser sessions of 43 self-hosted apps and 7 public APIs, replayed as MCP tools on AgentEnv.
50 replay environments REA built from app sessions
Note Start here: what these 44 environments are, how REA builds them, baseline results for two models, and a filterable gallery of all of them.
Note Developer & infrastructure · Gitea 28.1.0 (MIT), self-hosted on synthetic data. Git hosting, issues and code review. 77 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 11/12. Captured as a HAR.
Note Developer & infrastructure · Grafana 13.2.4 (AGPL-3.0), self-hosted on synthetic data. Observability dashboards, alerting and data sources. 48 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Developer & infrastructure · Sentry 26.9.0 (FSL-1.1-Apache-2.0), self-hosted on synthetic data. Error tracking and issue triage (application monitoring). 34 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Developer & infrastructure · HashiCorp Vault 1.21.4 (BUSL-1.1), self-hosted on synthetic data. Secrets and identity management (configuration store, policies, identity). 43 tools, 12 retrieval tasks; gpt-oss-120b 11/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Developer & infrastructure · HashiCorp Nomad 1.11.3 (BUSL-1.1), self-hosted on synthetic data. Workload orchestration (cluster scheduler). 39 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Developer & infrastructure · Redpanda Console 3.12.0 (BSL-1.1), self-hosted on synthetic data. Event streaming (Kafka-compatible topics, consumer groups and schema registry). 37 tools, 8 retrieval tasks; gpt-oss-120b 8/8, GLM-5.3-Flash 8/8. Captured as a HAR.
Note Developer & infrastructure · Airbyte 2.4.0 (Elastic License 2.0 (ELv2)), self-hosted on synthetic data. Data integration pipelines (ELT connectors, connections and sync jobs). 25 tools, 11 retrieval tasks; gpt-oss-120b 11/11, GLM-5.3-Flash 11/11. Captured as a HAR.
Note Developer & infrastructure · n8n 1.123.85 (Sustainable Use License 1.0), self-hosted on synthetic data. Workflow automation. 30 tools, 9 retrieval tasks; gpt-oss-120b 9/9, GLM-5.3-Flash 9/9. Captured as a HAR.
Note Developer & infrastructure · Keycloak 26.8.1 (Apache-2.0), self-hosted on synthetic data. Identity and access management (IAM). 40 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Developer & infrastructure · Healthchecks 4.4 (BSD-3-Clause), self-hosted on synthetic data. Cron job and heartbeat monitoring. 18 tools, 12 retrieval tasks; gpt-oss-120b 11/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Developer & infrastructure · Mailpit 1.31.6 (MIT), self-hosted on synthetic data. Email testing (SMTP capture and web mailbox). 15 tools, 11 retrieval tasks; gpt-oss-120b 11/11, GLM-5.3-Flash 11/11. Captured as a HAR.
Note Developer & infrastructure · NetBox 4.7.3 (Apache-2.0), self-hosted on synthetic data. Network source of truth (DCIM and IPAM). 42 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Developer & infrastructure · PocketBase 0.40.5 (MIT), self-hosted on synthetic data. Backend-as-a-service database and admin dashboard. 25 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Data & analytics · Metabase 0.64.2.8 (AGPL-3.0), self-hosted on synthetic data. Business intelligence and dashboards. 42 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Data & analytics · Umami 3.4 (MIT), self-hosted on synthetic data. Privacy-focused web analytics. 49 tools, 12 retrieval tasks; gpt-oss-120b 11/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Data & analytics · NocoDB 2026.09.1 (AGPL-3.0), self-hosted on synthetic data. No-code databases (spreadsheet-style tables). 28 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Data & analytics · Baserow 2.4.1 (MIT), self-hosted on synthetic data. No-code databases (spreadsheet-style tables). 31 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Data & analytics · Grist 1.7.21 (Apache-2.0), self-hosted on synthetic data. Spreadsheet-database (workspaces, documents, tables, records). 31 tools, 12 retrieval tasks; gpt-oss-120b 11/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Data & analytics · Directus 11.17.4 (BSL-1.1), self-hosted on synthetic data. Headless CMS and data platform. 43 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Work & publishing · Mattermost 12.0.0 (Team Edition, MIT-licensed server core), self-hosted on synthetic data. Team chat (channels, posts, teams). 37 tools, 12 retrieval tasks; gpt-oss-120b 11/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Work & publishing · Ghost 5.130 (MIT), self-hosted on synthetic data. Publishing and newsletters (CMS). 36 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Work & publishing · BookStack 26.09.1 (MIT), self-hosted on synthetic data. Team wiki and documentation (shelves, books, chapters, pages). 31 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Work & publishing · listmonk 6.2 (AGPL-3.0), self-hosted on synthetic data. Newsletters and mailing lists. 29 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Work & publishing · Vikunja 2.7 (AGPL-3.0-or-later), self-hosted on synthetic data. Task and project management (to-do lists, kanban). 28 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Work & publishing · EspoCRM 10.0.10 (AGPL-3.0), self-hosted on synthetic data. Customer relationship management (CRM). 88 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Work & publishing · Paperless-ngx 2.20 (GPL-3.0), self-hosted on synthetic data. Document management and archiving. 41 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Work & publishing · Memos 0.31.0 (MIT), self-hosted on synthetic data. Note-taking and microblogging (memos). 26 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Work & publishing · linkding 1.47.0 (MIT), self-hosted on synthetic data. Self-hosted bookmark manager. 23 tools, 12 retrieval tasks; gpt-oss-120b 11/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Work & publishing · Shlink 5.1.7 (MIT), self-hosted on synthetic data. URL shortening and visit analytics. 17 tools, 7 retrieval tasks; gpt-oss-120b 7/7, GLM-5.3-Flash 7/7. Captured as a HAR.
Note Work & publishing · Miniflux 2.3.3 (Apache-2.0), self-hosted on synthetic data. Feed reader (RSS and Atom). 22 tools, 7 retrieval tasks; gpt-oss-120b 7/7, GLM-5.3-Flash 7/7. Captured as a HAR.
Note Work & publishing · WordPress 7.1.3 (GPL-2.0-or-later), self-hosted on synthetic data. Blogging and content management (CMS). 21 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Photos & media · PhotoPrism 261007 (AGPL-3.0), self-hosted on synthetic data. Photo library and media management. 13 tools, 10 retrieval tasks; gpt-oss-120b 10/10, GLM-5.3-Flash 10/10. Captured by REA's capture-browser-scenario.
Note Photos & media · Lychee 6.10.4 (MIT), self-hosted on synthetic data. Photo management (self-hosted photo gallery). 15 tools, 12 retrieval tasks; gpt-oss-120b 11/12, GLM-5.3-Flash 12/12. Captured by REA's capture-browser-scenario.
Note Photos & media · Pinry 2.1.13 (BSD-2-Clause), self-hosted on synthetic data. Visual pin boards (tiling image bookmarks with boards and tags). 12 tools, 6 retrieval tasks; gpt-oss-120b 6/6, GLM-5.3-Flash 6/6. Captured by REA's capture-browser-scenario.
Note Photos & media · Immich 3.3.3 (AGPL-3.0), self-hosted on synthetic data. Self-hosted photo and video library. 45 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Photos & media · Jellyfin 10.10.7 (GPL-2.0), self-hosted on synthetic data. Media server and personal media library. 34 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Photos & media · Navidrome 0.64.2 (GPL-3.0), self-hosted on synthetic data. Self-hosted music streaming server. 25 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Photos & media · Komga 1.29.1 (MIT), self-hosted on synthetic data. Comics and ebook library server. 42 tools, 12 retrieval tasks; gpt-oss-120b 11/12, GLM-5.3-Flash 11/12. Captured as a HAR.
Note Photos & media · Sonarr 4.0.20.3014 (GPL-3.0), self-hosted on synthetic data. TV library management (profiles, custom formats, tags, connections). 48 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Home · Homebox 0.27.0 (AGPL-3.0), self-hosted on synthetic data. Home inventory and asset tracking. 28 tools, 10 retrieval tasks; gpt-oss-120b 10/10, GLM-5.3-Flash 10/10. Captured as a HAR.
Note Home · Grocy 4.7.1 (MIT), self-hosted on synthetic data. Household groceries, stock and chores (ERP). 45 tools, 12 retrieval tasks; gpt-oss-120b 11/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Home · Mealie 3.28 (AGPL-3.0), self-hosted on synthetic data. Recipe management and meal planning. 39 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Home · Tandoor Recipes 2.6.15 (AGPL-3.0 with Commons Clause), self-hosted on synthetic data. Recipe management and meal planning. 36 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Public data APIs · Art Institute of Chicago API v1 (CC0 1.0 (collection data)), public API, real records. Art museum collection (artworks, artists, galleries, exhibitions). 11 tools, 12 retrieval tasks; gpt-oss-120b 11/12, GLM-5.3-Flash 12/12. Captured by REA's capture-browser-scenario.
Note Public data APIs · Library of Congress JSON API (photos) v1 (U.S. public domain), public API, real records. Historical photograph collections (public API). 11 tools, 12 retrieval tasks; gpt-oss-120b 11/12, GLM-5.3-Flash 12/12. Captured by REA's capture-browser-scenario.
Note Public data APIs · openFDA v1 (U.S. public domain), public API, real records. Drug, device and food recalls, drug product listings and device classifications (U.S. government public API). 9 tools, 12 retrieval tasks; gpt-oss-120b 12/12, GLM-5.3-Flash 12/12. Captured as a HAR.
Note Public data APIs · USGS Earthquake Catalog 2.8.1 (U.S. public domain), public API, real records. Earthquake catalog and real-time seismic event feeds (geoscience). 10 tools, 7 retrieval tasks; gpt-oss-120b 7/7, GLM-5.3-Flash 7/7. Captured as a HAR.
Note Public data APIs · NOAA Tides and Currents API v1 (U.S. public domain), public API, real records. Coastal water levels, tide predictions and tide-station metadata (public data API). 12 tools, 9 retrieval tasks; gpt-oss-120b 9/9, GLM-5.3-Flash 9/9. Captured as a HAR.
Note Public data APIs · Federal Register API v1 (U.S. public domain), public API, real records. U.S. federal rules, proposed rules, notices and presidential documents. 10 tools, 10 retrieval tasks; gpt-oss-120b 9/10, GLM-5.3-Flash 9/10. Captured as a HAR.
Note Public data APIs · Treasury Fiscal Data API v1/v2 (U.S. public domain), public API, real records. Federal finance (public debt, exchange rates, Treasury statements, auctions and agency receivables). 16 tools, 12 retrieval tasks; gpt-oss-120b 8/12, GLM-5.3-Flash 12/12. Captured as a HAR.