/internal/docs/PRIORITIES.md

Priorities

The ranked work queue for the autonomous improvement loop. The architecture-review pass (the architect) owns this file: each run it turns the roadmap plus an internal scan (gaps in the services → agents → workflows lifecycle, API coherence, drift, tech debt, test and DX friction) into a single ordered list — highest-value first — and links each item to a tracking issue. The hourly continuous-improvement pass works the top item whose issue is still open. So the architect decides what, and the increment loop builds it.

Reading / editing. An item is done when its linked issue closes (the increment that builds it adds Closes #<issue>). Roadmap phase (Now → Next → Later) is the primary ordering; internal findings are interleaved by value, not kept in a separate list. The human can reorder this list — or the issues — at any time to redirect the loop; direction always wins.

Off-limits to the loop (the architect proposes these as notes, never as queue items the loop can auto-merge): brand/positioning copy, breaking public-API changes, architectural rewrites. Those go to the human.

Work queue (ranked)

1. Resume durable agent runs from checkpoints (#3401) — with the Now-phase getting-started contract closed by #3399, the highest-value Next-phase gap is making agent work survive restarts the way flows already do. This tightens the services → agents → workflows lifecycle around real, long-running work: checkpoint enough run state to resume safely, preserve guardrails/cancellation/deadlines, and prove completed tool calls are not duplicated.
2. Broaden end-to-end agent streaming coverage (#3402) — streaming is the next developer-visible seam after durability: provider tokens need to move consistently through ai.Stream, micro chat, Agent.Chat, and A2A streaming. CI-safe local coverage plus provider-gated conformance should lock down chunk ordering, terminal/error events, and fallback behavior.
3. Emit OpenTelemetry spans for agent runs (#3403) — once long runs can resume and stream, operators need the same run story in traces that developers see via micro runs: lifecycle boundaries, model/tool calls, approvals, retries, failures, and cancellation correlated by run ID without leaking sensitive payloads.

Seeded by Claude Code from the roadmap + open issues; thereafter maintained by the architecture-review pass.