fix: bypass broken A2A gateway, call Zhipu LLM directly

fix: delete devices dir to force fresh auto-pair, skip error replies

fix: patch paired device scopes for OpenClaw 2026.2.19+ compatibility

fix: clean up invalid auth.defaultScope from persisted config

fix: revert invalid OpenClaw config keys causing RUNTIME_ERROR

fix: add operator scope to gateway auth and A2A security config

feat: bilingual conversation, word-wrap bubbles, remove unused panels

fix: switch conversation to English, longer replies (2-3 sentences), wider bubbles

fix: clean up conversation prompt to avoid prefix leakage

feat: chat log panel + improved conversation orchestrator

feat: add conversation bubble display and A2A chat orchestrator

fix: reduce verbose logging to prevent HF Spaces log viewer crashes

fix: restore 302 redirect for token so frontend JS can read it from URL

fix: apply iframe CSP fix before URL rewrite to prevent frame-ancestors 'none'

fix: use server-side URL rewrite instead of 302 redirect for token injection

Upload scripts/token-redirect.cjs with huggingface_hub

Upload scripts/sync_hf.py with huggingface_hub

Upload scripts/entrypoint.sh with huggingface_hub

Upload scripts/token-redirect.cjs with huggingface_hub

Upload scripts/a2a-proxy.cjs with huggingface_hub

Upload scripts/a2a-proxy.cjs with huggingface_hub

Upload scripts/entrypoint.sh with huggingface_hub

Upload scripts/sync_hf.py with huggingface_hub

Upload scripts/a2a-proxy.cjs with huggingface_hub

Upload folder using huggingface_hub

feat: serve pixel office animation at / with static asset serving

fix: accept 302 redirect as healthy in OpenClaw health check

feat: add remote agent monitoring and A2A activity tracking to proxy

feat: add A2A gateway support (activated by A2A_PEERS env var)

Revert "feat: add pixel animation homepage with Express proxy layer"

feat: add pixel animation homepage with Express proxy layer

fix: switch Zhipu provider to Anthropic-compatible endpoint

fix: update Zhipu model IDs in default config and dataset template

fix: update Zhipu AI model IDs to current API lineup

fix: add required models array to zhipu provider config

fix: rebuild providers from scratch on each startup

feat: add Zhipu AI (z.ai) as a native LLM provider

feat: auto-fill gateway token via URL redirect

fix: remove tao-shen/HuggingClaw-data refs to prevent unwanted linked spaces

test: revert dataset derivation to observe linked spaces behavior

fix: restore OPENCLAW_DATASET_REPO priority over SPACE_ID derivation

fix: always derive dataset from SPACE_ID, ignore OPENCLAW_DATASET_REPO on HF

fix: revert to token auth with default "huggingclaw"

feat: default to no-auth mode — Control UI connects instantly

fix: auto-derive dataset name from SPACE_ID, not HF_TOKEN username

feat: add GATEWAY_TOKEN env var (default: huggingclaw)

fix: restore dangerouslyDisableDeviceAuth for HF Spaces

config: set hardcoded gateway token "huggingclaw" as default

remove: drop gateway password/token auth entirely — device auth only

security: enforce device auth — only paired browsers can access Control UI