Fetching metadata from the HF Docker repository...
-
auto-fallback: try InferencePort, if 429 use KeylessAI
ausername-12345
-
switch to InferencePort direct API
ausername-12345
-
back to KeylessAI (completions.me 401)
ausername-12345
-
read API key from env var with fallback
ausername-12345
-
switch to completions.me with gpt-5.2
ausername-12345
-
compact history to 20 messages: older ones get summarized into context
ausername-12345
-
log all requests with full details
ausername-12345
-
switch to KeylessAI (no key, no limits, no signup)
ausername-12345
-
add web_type tool for filling forms and typing in textboxes
ausername-12345
-
add web_search, web_click tools for interactive browsing
ausername-12345
-
add web browsing: web_fetch tool + tool call loop
ausername-12345
-
restore original request parsing for shortcuts + history
ausername-12345
-
fix shortcuts empty history, add debug logging
ausername-12345
-
add GET /history, handle messages array in shortcuts
ausername-12345
-
separate POST /reset endpoint for clearing history
ausername-12345
-
shortcuts maintains conversation history, reset with {"reset":true}
ausername-12345
-
rename relay_history to history
ausername-12345
-
restore relay_history as string
ausername-12345
-
remove relay_history from response
ausername-12345
-
relay_history as string
ausername-12345
-
relay_history excludes system prompt; API still sends it
ausername-12345
-
shortcuts: send only messages, no system prompt
ausername-12345
-
shortcuts: send all messages without trimming
ausername-12345
-
add relay_history to response
ausername-12345
-
add request logging on failure (method, url, headers, body)
ausername-12345
-
restore default system prompt for /shortcuts
ausername-12345
-
make /shortcuts accept simpler formats (text/message)
ausername-12345
-
add POST /shortcuts endpoint (alias for /relay)
ausername-12345
-
use lightning model (Lightning-Text v2)
ausername-12345
-
fix: use llama-3.1-8b-instant model (llama3-8b-8192 decommissioned)
ausername-12345
-
sync with sebebeb-relay.hf.space: fix API URL to /v1/chat/completions, add debug endpoint, restore OpenAI payload format
ausername-12345