Skip to content

feat(pi): orchestrator runTask — per-task pi sessions with in-process queue tools#853

Draft
gewenyu99 wants to merge 5 commits into
mainfrom
experiment/orchestrator-pi-runtask
Draft

feat(pi): orchestrator runTask — per-task pi sessions with in-process queue tools#853
gewenyu99 wants to merge 5 commits into
mainfrom
experiment/orchestrator-pi-runtask

Conversation

@gewenyu99

@gewenyu99 gewenyu99 commented Jul 10, 2026

Copy link
Copy Markdown
Collaborator

Implements runTask on the pi harness so the orchestrator runs model-agnostic through pi — pairs with PostHog/context-mill#181 (the flow-scoped agents + step-skills it consumes).

Testing

  • check out this branch and the context-mill PR branch (experiment/orchestrator)
  • serve context-mill locally: PORT=8770 npm run dev in the context-mill checkout
  • point REMOTE_SKILLS_BASE_URL in src/lib/constants.ts at http://localhost:8770 (revert after, it breaks the health-check unit tests while edited)
  • copy a fixture app somewhere disposable, e.g. wizard-workbench express-todo, without node_modules
  • run from the wizard checkout:
    SNAP_OUT=/tmp/snaps APP_DIR=<app copy> POSTHOG_KEY_FILE=<personal api key file> PROJECT_ID=<project id> \
    WIZARD_CI_FLAG_OVERRIDES='{"wizard-orchestrator":"true","wizard-use-pi-harness":"true"}' \
    npx tsx scripts/tui-snapshots.no-jest.ts
    
  • expect in /tmp/posthog-wizard.log: harness=pi (flag) and sequence=orchestrator (flag), then the queue draining
  • expect at the end: a 7/8 or 8/8 outro, posthog-setup-report.md in the app, a real dashboard in the project
Full TUI run — 24 frames (7/8 done, 1 reasoned skip, 0 failed, 8m01s, dashboard 1825791)

01-intro — intro — orchestrator + pi flags resolved

01-intro

02-auth — auth

02-auth

03-run — run start

03-run

04-run — seed planning

04-run

05-run — queue seeded, install running

05-run

06-run — install

06-run

07-run — install done

07-run

08-run — init running

08-run

09-run — init

09-run

10-run — init done

10-run

11-run — parallel branch starts

11-run

12-run — identify + capture + error-tracking in parallel

12-run

13-run — instrumentation

13-run

14-run — identify skipped with reason

14-run

15-run — instrumentation done

15-run

16-run — build — real npm install

16-run

17-run — build verifying

17-run

18-run — dashboard task

18-run

19-run — dashboard insights via posthog_exec

19-run

20-run — report writing from handoffs

20-run

21-outro — outro

21-outro

22-outro — outro — 7/8 steps, report on disk

22-outro

23-mcp — mcp screen

23-mcp

24-slack-connect — slack connect

24-slack-connect

🤖 Generated with Claude Code

releaser-wizard Bot and others added 5 commits July 8, 2026 16:30
… queue tools, capability clamp

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…ting, framework-variant step-skill resolution

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…or-pi-runtask

# Conflicts:
#	CHANGELOG.md
#	package.json
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant