Headroom sidecar
Colibri can optionally ask a local headroom-ai sidecar to compress tool results
before they are rendered through the session read path.
This is off by default. The daemon does not start or supervise Headroom; it only connects to an already-running Unix socket when explicitly enabled.
Configuration
Section titled “Configuration”COLIBRI_HEADROOM_ENABLED=1COLIBRI_HEADROOM_SOCKET=/var/run/colibri/headroom.sock # optional defaultStart the sidecar with a Python environment that can import headroom:
python3 scripts/headroom-sidecar.py --socket /var/run/colibri/headroom.sockThen start colibri-daemon with the environment above. Check connection state
with colibri status; the status JSON includes:
{ "headroom": { "enabled": true, "socket": "/var/run/colibri/headroom.sock", "connected": true }}Protocol
Section titled “Protocol”The daemon keeps one Unix-socket connection open and sends newline-delimited JSON requests. The sidecar must support multiple requests on that same accepted connection.
Request:
{"id":"tool-name","raw":"tool output text","role":"tool"}Response:
{"id":"tool-name","compressed":"shorter text","tokens_before":100,"tokens_after":25}If tokens_after >= tokens_before, Colibri keeps the original tool result. If
the sidecar errors, disconnects, or times out, Colibri degrades to the original
uncompressed content.
Current daemon timeout: 5 seconds per sidecar request.
Scope and caveats
Section titled “Scope and caveats”- Compression is opt-in and best-effort.
- The sidecar currently affects session prompt/message rendering paths that pass
a
HeadroomSidecar; it is not a global daemon-wide replacement for existing cost-mode compaction. - Keep
COLIBRI_HEADROOM_ENABLED=0for ISO/live-USB defaults unless Headroom is staged and validated on the target image. FreeBSD: ONNX/ORT-backed extras need local packaging; use a known-good Python environment and validatescripts/headroom-sidecar.pydirectly before enabling the daemon flag.
See also
Section titled “See also”- cost-model — cost discipline and when to use Headroom
- operator-cli — status JSON reports headroom connection state