/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. Add cross-provider agent conformance harness (#3491) — the verification loop shipped in #3489, so the next highest-value Now-phase gap is confidence that the same services → agents → workflows contract works across providers. The README and website sell one harness with pluggable providers, while the roadmap calls out that each provider has its own tool-call loop; a shared no-secret/live-key conformance scenario is the fastest way to prevent provider drift from undermining the 0→hero path.

2. Harden agent loop failure and cancellation semantics (#3492) — once provider behavior is comparable, the remaining Now-phase operational risk is whether real runs fail safely: timeouts, rate limits, cancellation, deadline propagation, retry/backoff, and inspectable terminal state. This keeps the harness dependable under production conditions without adding a new product surface.

3. Add a scheduled agent run harness contract (#3486) — scheduled, looping, work-performing agents remain the right Next-phase cohesion target after hardening. This should compose existing services, agents, flows, store-backed memory, verification, and run inspection so the scaffold → run → chat → inspect lifecycle extends to unattended work without adding a hosted scheduler or breaking public APIs.

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