OpenClaw v2026.11.3

OpenClaw v2026.11.3 is a security-focused release of the open-source personal agent: private subagent results stay out of streamed replies, unanswered tool failures are now reported instead of ending a turn as a success, and plugin and search credentials are kept scoped to their own connections.

Why it matters

People running OpenClaw as an always-on personal agent hand it many API keys and channels. This release tightens where those credentials can travel and makes failed tool calls visible instead of silent.

What's new

• Private subagent completions are kept out of streamed chat replies.
• A tool failure that never got an answer is now reported, rather than the turn quietly ending as a success.
• Provider authentication that stalls is stopped at the run deadline, and cancellation is passed through authentication work.
• Cached plugin results, Firecrawl search keys and embedding credentials stay tied to their own connection or endpoint, and inline environment credentials are redacted from configuration reads.
• Managed Signal daemons and bundled language servers no longer inherit unrelated Gateway secrets; the release notes say there are no new configuration options.

Taken from the project's own release notes. Claims about speed or accuracy are the maker's unless stated otherwise.

Quick facts

Links

• Official site: https://openclaw.ai
• Repository: https://github.com/openclaw/openclaw
• Release notes: https://github.com/openclaw/openclaw/releases/tag/v2026.11.3
• Docs: https://docs.openclaw.ai

Related ranking

See how established tools compare in Best AI Agents That Do Tasks for You. New releases aren't scored until we review them against the rubric.

What we couldn't verify

• quick_facts.platform (supported OSes not restated in these release notes)

Checked 2026-11-11 from the release notes's own pages. Prices and star counts change, so confirm on the official site. Listing here is not a ranking or an endorsement.