feat(orchestrator): ci-excluded task types#639
Merged
gewenyu99 merged 3 commits intoJun 18, 2026
Merged
Conversation
🧙 Wizard CIRun the Wizard CI and test your changes against wizard-workbench example apps by replying with a GitHub comment using one of the following commands: Test all apps:
Test all apps in a directory:
Test an individual app:
Show more apps
Results will be posted here when complete. |
This was referenced Jun 10, 2026
Collaborator
Author
This was referenced Jun 10, 2026
3c02336 to
94e28b9
Compare
2365eef to
235e1d2
Compare
This was referenced Jun 16, 2026
94e28b9 to
bd393b7
Compare
235e1d2 to
6b8b1f7
Compare
edwinyjlim
approved these changes
Jun 17, 2026
6b8b1f7 to
e116168
Compare
Base automatically changed from
experiment/orchestrator-telemetry
to
experiment/orchestrator-01-bootstrap-gating
June 18, 2026 14:23
The CI harness can exclude task types from a run via a CI-build-only env var (a comma-separated list, e.g. dashboard — CI runs must not create dashboards). Excluded types do not exist for the run: the seed cannot enqueue them and no agent is spun up. The policy lives with the harness that owns it; the wizard and the served content stay run-mode agnostic. Published builds strip the path, smoke-enforced alongside the flag overrides. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The server reads the user's region from the bearer token, so the EU subdomain (a Claude Code OAuth workaround) is no longer needed. Removes the host-parsing branch in both the orchestrator bootstrap and the linear MCP runner. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Review follow-ups on the queue and loader: - TaskStatus 'in_progress' -> 'running' (drop the snake_case outlier) - inline the nowIso() one-liner - document that dependsOn is a DAG by construction (ids point only at earlier tasks), so cycles cannot form - cap the queue at 30 tasks as a runaway backstop; real sizing rests on agent/skill design - move agent-prompt-loader from programs/orchestrator/ to lib/agent/ Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
bd393b7 to
7d45f09
Compare
344790b
into
experiment/orchestrator-01-bootstrap-gating
10 checks passed
gewenyu99
added a commit
that referenced
this pull request
Jun 18, 2026
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
gewenyu99
added a commit
that referenced
this pull request
Jun 18, 2026
…rd-orchestrator flag) (#606) Experimental task-queue orchestrator, collapsed from an 11-PR stack into this single branch. Behind the `wizard-orchestrator` feature flag, off by default — with the flag off it is a no-op and the existing linear flow runs unchanged. ## Wizard PRs folded in (squash-merged into this branch) - #607 — queue persistence - #608 — orchestrator MCP tools: enqueue_task / complete_task / read_handoffs, with guards - #609 — executor - #619 — agent loader - #620 — email targeting - #635 — CI flag overrides - #637 — ephemeral skills - #638 — telemetry (responsiveness A/B spine) - #639 — CI task filter - #677 — run cache + responsiveness analytics - #678 — CI health-check advisory ## Related PRs, other repos - PostHog/context-mill#181 — `agents` content type: agent prompts plus the `basic-integration` step-skills, built and served alongside skills - PostHog/wizard-workbench#1879 — CI flag-override and task-exclude dispatch inputs (dev/CI stack) ## Flag gating — why this is a no-op with the flag off - `runProgram` forks on `isOrchestratorEnabled(flags)` = `flags['wizard-orchestrator'] === 'true'` (default false). Off, it runs `runLinearProgram`, the existing path, untouched. - The shared `runAgent` threads the same flag through, so its orchestrator-specific message handling is gated too. - The CI flag-override (`WIZARD_CI_FLAG_OVERRIDES`) is guarded by `NODE_ENV === 'production'` and stripped from published builds — a smoke test asserts the env-var names are physically absent — so it cannot flip the flag for real users.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.

Basically, the agent wastes a lot of time and tokens in CI trying to create dashboards. We don't let it, it'll be missing scopes. That's fine, this just skips that step in queue.
The CI harness can exclude task types from a run.
WIZARD_CI_EXCLUDE_TASKSis a comma-separated list, the registry drops those types, the seed cannot enqueue them, and no agent is ever spun up for them. The policy lives with the harness that owns it (the workbench workflow setsdashboard, CI runs must not create dashboards), so the wizard and the served agent content stay run-mode agnostic. CI-build only, same dead-strip and smoke enforcement as the flag overrides. The seed prompt plans around missing types, soreportfollowsbuildwhendashboardis absent.