Skip to content

LiveContext CE v0.1.9

Choose a tag to compare

@livecontext-ai livecontext-ai released this 04 Jul 19:04

What is new

v0.1.9

Browser agent

  • The browser agent now relays its LLM through your cloud connection when the install is cloud-linked (like the chat and workflow agents), so agent_browse no longer needs a separate direct provider key. Full token/cost/model observability is recorded per run.
  • Fixed web_search: added the missing SearXNG sidecar to the browser-agent profile (it was returning no results).
  • Removed the "Vision only" filter on the Browser Agent model picker; any configured model can drive the browser.

Models and bundles

  • Manual "Sync now" for the model catalog is now asynchronous and survives page navigation, with a Stop control.
  • Seed-inserted models get default categories and orphaned models are repaired, so newly added models are selectable in the pickers.
  • The CE model-catalog seed is version-gated and refreshed per release (partial merge that never clobbers enrichment).

Fixes and docs

  • Clearer bridge setup instructions (run the bridge from the project root).
  • Removed the misleading "needs a Google/Gemini key" note from the browser-agent docs.

Update an existing install

git pull
docker compose pull
docker compose up -d

The compose file pins this release's image tag, so run git pull first. Your data lives in Docker volumes and is kept across updates.

Fresh install

git clone https://github.com/livecontext-ai/livecontext-ce
cd livecontext-ce
docker compose up -d

Then open http://localhost:3000

Images

  • ghcr.io/livecontext-ai/livecontext-ce:v0.1.9
  • ghcr.io/livecontext-ai/livecontext-ce-frontend:v0.1.9
  • ghcr.io/livecontext-ai/livecontext-ce-bridge:v0.1.9
  • ghcr.io/livecontext-ai/livecontext-ce-screenshot-renderer:v0.1.9 (opt-in renderer profile)