Roland f86058ee8a ai: the /api/chat agent loop — SSE streaming, modes, tool execution ... Tool definitions are generated in-process from the /api/ai blueprint's own OpenAPI operations (operationId = tool name, description = tool description, dataclass bodies + declared parameters = one flat JSON schema) — the spec IS the catalog, nothing to drift. Manage endpoints never become tools. A mode is a server-enforced system prompt + tool allowlist (Help / Research / System per ai.md); the executor can only reach allowlisted /api/ai endpoints, path parameters are encoded and traversal-rejected so a crafted argument cannot escape an endpoint. POST /api/chat streams SSE (status/token/tool_call/tool_result/done/ error), starts llama-server on demand, validates tool args server-side with instructive errors the model can retry on, caps tool results with a paging hint, and bounds the loop by ai.yaml max_tool_calls plus a hard timeout. Caddy passes the stream unbuffered (flush_interval -1) in both deployments. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>		2026-09-24 14:09:54 -06:00
..
Caddyfile	ai: the /api/chat agent loop — SSE streaming, modes, tool execution	2026-09-24 14:09:54 -06:00
Dockerfile	Map previews that never drew, and a Docker map with no world under it (#658)	2026-08-11 20:37:23 -06:00
entrypoint.sh	Replace nginx with Caddy for TLS termination and reverse proxying.	2026-03-07 18:48:44 -07:00