RepoJournal
Agentic Coding Agentic Coding
123 wires and counting

$ follow Agentic Coding

Keep up with Agentic Coding in about 3 minutes: what actually shipped — the commits, pull requests, releases, and security advisories that matter.

or

fair warning: these emails are deeply technical. diffs, version numbers, CVEs, benchmark deltas. if that's not your idea of a good read, this isn't your newsletter.

Folds into your digest — weekly by default, monthly if you prefer. Unsubscribe in one click.

$ status

wire 2026-11-16
stories 103

© 2026 RepoJournal Home Showcase How it works Privacy

$ the-wire · showcase

codex exec goes remote, opencode answers crashes with readable 500s

By RepoJournal · Filed · About Agentic Coding · Composed from the cited sources · methodology

Codex and opencode both shipped operator-visible changes today: remote app servers in `codex exec`, and a server that explains its 500s instead of returning an empty body.

Support remote app servers in `codex exec` openai/codex

by copyberry

`codex exec --remote <ADDR>` now runs a noninteractive turn against an existing app server's configuration and credentials, starting a new thread or resuming one by UUID, with human-readable or JSON output and optional bearer-token auth via `--remote-auth-token-env`. Remote options are accepted before or after `exec`, and conflicting endpoints are rejected, so CI jobs can share one running serv...

fix(server): answer handler defects with a readable 500 instead of an empty one anomalyco/opencode

by rekram1-node

A request that crashes in a handler used to return a 500 with an empty body; it now carries JSON naming storage as the cause when that is the problem, and otherwise a ref to find the crash in the server log. Client code that branched on an empty 500 body needs to read the JSON instead, and messages never name the database engine.

feat(ai): scope HTTP middleware to any call, for every modality anomalyco/opencode

by rekram1-node

Plugin HTTP hooks previously ran only on chat calls; the `@opencode/ai` half of the change scopes middleware to image, video, decisions, embeddings, speech and transcription calls plus polling and downloads. Core wiring comes in a follow-up, so hooks that gate non-chat model traffic will not fully take effect until that lands.

fix(oi-kit): bound empty response retries aaif-goose/goose

by DOsinga

Empty provider responses now recover through a bounded state-machine transition, with agent-only retry messages capping recovery at three retries, and retry counting ignores request-preparation context messages so each attempt still passes preparation and turn-budget checks. The point for operators is that a silently empty response no longer hangs or loops the turn.

fix(databricks): drop thinking blocks bound to a different prefix aaif-goose/goose

by baxen

A Claude Fable 5.2 session on the Databricks AI Gateway used to fail permanently with a 400 signature error once its system prompt changed; thinking blocks bound to a different prefix are now dropped. That removes the dead end for anyone editing a system prompt mid-session against that gateway. Elsewhere, OpenHands synced `agent_settings.agent_kind` with the active profile on cloud and retained...

Quick answers

What shipped in Agentic Coding on November 16, 2026?
Codex and opencode both shipped operator-visible changes today: remote app servers in `codex exec`, and a server that explains its 500s instead of returning an empty body. In total, 50 commits, 50 pull requests, and 3 releases landed.
Who contributed to Agentic Coding on November 16, 2026?
9 developers shipped this update, including GitHub Actions (publish-mods), copyberry, rekram1-node, thdxr, stefanielizeth73-art, suzhiguo7, dependabot, and DOsinga, and 1 more.
What were the notable Agentic Coding updates?
Support remote app servers in `codex exec`, fix(server): answer handler defects with a readable 500 instead of an empty one, and feat(ai): scope HTTP middleware to any call, for every modality.