Skip to content

Refresh planner priorities after retry hardening #4704

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
asim merged 1 commit into from
Jul 11, 2026
Merged

Refresh planner priorities after retry hardening #4704

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions .github/loop/PRIORITIES.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,8 +21,8 @@ changes, architectural rewrites. Those go to the human.

## Work queue (ranked)

1. **Harden agent provider failure resilience** ([#4650](https://github.com/micro/go-micro/issues/4650)) — With scheduled cross-provider conformance closed by #4678, the first-agent fixture races closed by #4684/#4667, Nack redelivery fixed by #4694, and initial provider failure metadata recorded by #4688, the next Now-roadmap reliability seam is making timeouts, cancellation, rate limits, retry/backoff, and inspectable failure metadata predictable through the agent loop. Keep this scoped to existing behavior and tests; surface any breaking API or default changes as human review notes instead of queue work.
2. **Verify first-agent docs commands in CI** ([#4696](https://github.com/micro/go-micro/issues/4696)) — The North Star says developer adoption is the current goal, and the README now offers a long walkable path (`micro agent demo`, `quickcheck`/`debug`, `examples`, `zero-to-hero`, docs, and no-secret examples). Turn those breadcrumbs into a focused no-secret command/documentation contract so the 0→1 and 0→hero on-ramp stays runnable instead of becoming aspirational copy while internal hardening continues.
1. **Verify first-agent docs commands in CI** ([#4696](https://github.com/micro/go-micro/issues/4696)) — With #4702 closing the broad provider retry-cancellation seam, the highest-value Now-roadmap work returns to developer adoption: make the README/website first-agent path a focused no-secret command contract. The mission is to make services → agents → workflows feel like one runtime, and the current risk is not lack of depth but an on-ramp that can drift from runnable behavior. Keep this scoped to verified commands, example wayfinding, and short mock-model checks; do not redesign public APIs.
2. **Harden AtlasCloud plan-delegate tool-call recovery** ([#4699](https://github.com/micro/go-micro/issues/4699)) — Recent provider resilience work shipped retry cancellation and failure metadata, but the live AtlasCloud plan/delegate harness exposed a concrete malformed/nested tool-call recovery gap. Fixing it protects the agent loop's operability across providers while keeping the queue honest about current CI signals. Keep this to recovery/rejection behavior and regression coverage; surface any provider-contract or default-behavior changes for human review.

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