/.github/loop/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. Update AI capability tests for minimax provider registration (#3787) — the latest master Run Tests workflow is red because the AI provider capability fixtures still omit the newly registered minimax provider. This is the top Now-phase item because the autonomous loop's evaluator is blocked by a deterministic unit-test regression, and the fix is narrow, CI-verifiable, and directly protects cross-provider adoption work.
2. Fix atlascloud plan-delegate harness missing delegated notify side effect (#3760) — PR #3783 closed the buyer-recipient universe blocker (#3772), leaving plan/delegate as the live atlascloud/minimax 0→hero services → agents → workflows failure. Keep this ahead of broader hardening until the conductor → delegated comms → notify path reliably proves the multi-agent side effect developers see in the harness.
3. Isolate file-store tests from shared default directory (#3751) — repeated go test -race -cover ./... failures still point at file-store reliability under the shared default directory/table setup. This remains the highest-value store/CI stability issue because a flaky evaluator erodes the loop's ability to ship adoption work safely.
4. Stabilize file-store suffix expiry test timing (#3780) — the newer store failure is a narrower timing-sensitive suffix-expiry assertion under -race -cover. Keep it adjacent to #3751 but separate because it may need a focused TTL/assertion fix even if directory isolation improves the broader file-store tests.
5. Propagate agent run cancellation and deadlines through model and tool calls (#3544) — once the live harness and red CI blockers are cleared, the highest-value remaining resilience gap is predictable failure semantics across agent runs, model calls, tool calls, plan/delegate, and flow handoffs. Tool retries, live-provider deadline tuning, delegated-plan completion, and side-effect enforcement are in place; the lifecycle still needs cancellation/deadline propagation so work fails safely instead of becoming opaque loops.
6. Emit OpenTelemetry spans for agent run timelines (#3525) — recent work made runs inspectable, correlated trace metadata through scheduled dispatch, verified restart resume, added opt-in tool retries, hardened provider conformance, and fixed provider-emitted text tool calls. The next Next-phase step is to turn that RunInfo foundation into standard OTel spans for agent runs, model calls, tool calls, checkpoint/resume, cancellation/deadlines, and failures.
7. Add an AP2 mandate layer over A2A and x402 (#3552) — this is a forward interop investment, not a Now-phase blocker: Go Micro already has A2A agents and x402 paid tools, so a small signed-mandate foundation can keep agent payments aligned with the open-protocol story without pulling the queue away from adoption, resilience, or observability. Keep it additive and opt-in while the AP2/FIDO work settles.

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