Repository navigation
id: --json transcript output in world line mode #563
Description
What and why
Deterministic machine-readable replies in line mode, so scripted round-trips against a running world need no terminal; useful for smoke tests and fixtures. Agents and CI currently have to scrape terminal text to script a world.
Source
- RPT §6 'Agents and CI'
- RPT §7 Step 3.3
- RPT §12 Later
RPT = doc/2026-10-18T18-40-00Z_report_platform_world_state/2026-10-18T18-40-00Z_report_platform_world_state.md, REF = doc/2026-10-18T00-00-00Z_reference_platform_state/2026-10-18T00-00-00Z_reference_platform_state.md. Both are uncommitted reports in the platform-world-apps-migration worktree.
Current status
Not started. World line mode prints terminal text only (pkgs/id/src/world_ssh.rs:525-600 send). The --json flags in pkgs/id/src/cli.rs (:1047-1079) belong to group/doc commands, not worlds.
Acceptance criteria
- Line mode can emit deterministic machine-readable replies (
--json) - A smoke test drives a running world through it with no terminal
- The output schema is documented
Size
M
Prerequisites
- id: one line grammar for SSH command mode, the
id world joinpump and CLI verbs (--command-mode) #546 — D1-041 One line grammar shared by SSH command mode,id world joinpump and CLI verbs (+--command-mode)
Metadata
Metadata
Assignees
Labels
Package/app: id (pkgs/ or apps/ in this repo)Medium: expected under 1 day of workNew feature or requestNew feature or capability requestIntake review done: acceptance criteria, links, and normalized labels addedFully doable remotely: design, config, code, cloudOrdinary priority (the default)Filed directly by a human
Activity
dezren39 commentedon Oct 24, 2026
id train triage (2026-10-24): too-big-needs-design — deterministic --json transcript shape for world line mode.
dezren39 commentedon Oct 26, 2026
Intake review
Fleet-monitor intake pass (2026-10-26). Body already carries a full
## Acceptance criteriasection, so this pass is label normalization plus a liveness check — no body edit.Verified live on main
e3467560(2026-10-26): no--jsontranscript path exists for world line mode. The JSON inpkgs/id/src/world_ssh.rsis internal frame encoding (input/chatframes), not a user-facing transcript format, and the--jsonflags incli.rsbelong to group/doc commands. (The transcript contract atcli.rs:1785covers records verification, a different surface.)Related: prerequisite #546 (one shared line grammar) is still open and unenriched — the output schema should be defined against that grammar, not invented independently. Sibling terminal-output work: #562, #561.
Risks / notes: schema stability is the real decision here; once scripts consume the format it is a contract. Deterministic ordering and stable field names should be acceptance-tested, not just eyeballed.
Suggested approach: define the reply schema alongside #546's grammar, emit it behind
--jsonin line mode only, and add the no-terminal smoke test from the acceptance criteria in the same PR.144 remaining items
dezren39 commentedon Nov 2, 2026
🤖 Agent note · parked — Drew ownership stop remains
--jsontranscript output in world line mode #563; no work performed.--jsontranscript output in world line mode #563 is OPEN/Drew-assigned and canonical PR feat(id): add JSON transcript for world command mode (#563) #3776 remains OPEN, non-draft, Drew-authored/assigned. Existing hard stop remains; no PR internals or worktree inspected.--jsontranscript output in world line mode #563; PR feat(id): add JSON transcript for world command mode (#563) #3776; queue task 4838dezren39 commentedon Nov 2, 2026
agent-note:v1 — holder: repo-worker-code (OpenChamber scheduled lane), at 2026-11-02T13:50:53Z. event: picked-up; task: 4838; queue: repo-health-code-pilot.
dezren39 commentedon Nov 2, 2026
agent-note:v1 — holder: repo-worker-code (OpenChamber scheduled lane), at 2026-11-02T13:51:01Z. event: parked; task: 4838; queue: repo-health-code-pilot. Canonical PR #3776 remains open and Drew-owned; hard stop. Resume after it closes/merges or explicit ownership release, then revalidate fresh. PR internals, checks, reviews, branch, worktree and source were not inspected or touched.
dezren39 commentedon Nov 2, 2026
🤖 Agent note · picked up
--jsontranscript output in world line mode #563 — PR feat(id): add JSON transcript for world command mode (#563) #3776 is on this task's own fleet branch queue/4838-563-deterministic-json-world-line-transcripts and its body binds it to issue id:--jsontranscript output in world line mode #563, so the prior Drew-owned park was the dezren39 shared-account false positive (same pattern corrected on task 5693 / PR fix(id): widen base+U+FE0F only for emoji-presentable bases (#3265) #4355 this shift). No held / user/owned / do-not-merge / do-not-auto-edit labels on issue id:--jsontranscript output in world line mode #563 or PR feat(id): add JSON transcript for world command mode (#563) #3776. Revalidating PR feat(id): add JSON transcript for world command mode (#563) #3776 head, checks, and Codex findings fresh; the resume point is the earlier targeted_tests_failed checkpoint, so test state gets verified before any merge.Merge pull request #3776 from developing-today/queue/4838-563-determi…
dezren39 commentedon Nov 2, 2026
🤖 Agent note · done
--jsontranscript output in world line mode #563 closed by the merge. The repeated Drew-owned parks on this task were the dezren39 shared-account false positive: PR feat(id): add JSON transcript for world command mode (#563) #3776 sits on task 4838's own fleet branch queue/4838-563-deterministic-json-world-line-transcripts, its body binds it to issue id:--jsontranscript output in world line mode #563, and no hard-stop label exists on either item. Verification this shift: Codex re-review requested and completed clean on the current head 5e4535e (no major issues, zero inline findings); the one commit after the originally reviewed head is a one-line PUMP_HELP string preservation, read directly; PR-body test evidence (schema + CLI/unit/live-world regressions) stands.