A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

5 stars 5 forks 5 watchers Go Apache License 2.0
agent-infrastructure ai-agents browser-automation desktop local-llm mcp mcp-server model-context-protocol ollama openclaw playwright
81 Open Issues Need Help Last updated: Jun 22, 2026

Open Issues Need Help

View All on GitHub

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
good first issue docs contributor-experience priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P3 pack-candidate

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
help wanted good first issue

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement good first issue pack-candidate

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 priority/P2

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue Phase 8 priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
good first issue

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 marketplace priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 integrations/hermes-agent

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 integrations/gemini-cli

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 integrations/claude-desktop

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 integrations/claude-code

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P1 pack-candidate

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P3 pack-candidate

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P3 pack-candidate

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P3 pack-candidate

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P3 pack-candidate

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P3 pack-candidate

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P3 pack-candidate

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P3 pack-candidate

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 pack-family/language

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 pack-family/research

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 pack-family/http

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 pack-family/shell

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 pack-family/fs

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 pack-family/vision

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 pack-family/desktop

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 pack-family/doc

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 pack-family/slides

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 pack-family/github

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 pack-family/repo

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
documentation good first issue priority/P2 pack-family/web

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
good first issue docs contributor-experience priority/P2

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 marketplace priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 marketplace priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
good first issue Phase 8 contributor-experience marketplace priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 marketplace priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 marketplace priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 marketplace priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
good first issue Phase 8 docs priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement good first issue Phase 8 priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted Phase 8 priority/P3

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted priority/P2

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement good first issue priority/P2

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
good first issue docs contributor-experience priority/P2

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement good first issue contributor-experience priority/P2

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement good first issue provider priority/P2

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement good first issue provider priority/P2

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement help wanted priority/P1

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement good first issue provider

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright
enhancement good first issue provider

A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.

Go
#agent-infrastructure#ai-agents#browser-automation#desktop#local-llm#mcp#mcp-server#model-context-protocol#ollama#openclaw#playwright