Brummel
5c3bd9ab24
Claude Code categorically forbids subagents from spawning other subagents (code.claude.com/docs/en/sub-agents and the Agent SDK subagents page). The `or.1` architecture (and `pr.1` that ran on top of it) presumed a named-exception for `ailang-implement- orchestrator` to dispatch implementer / spec-reviewer / quality- reviewer / tester per task. That exception never existed at the platform level; the `Agent` tool was silently dropped from the orchestrator-agent's tool set at dispatch time. Architecture revised, doc-only: - Per-task phases (implementer → spec-compliance check → quality check) now run as sequential role-switches in the orchestrator- agent's own context, not as nested subagents. - The four role-files (implementer / spec-reviewer / quality- reviewer / tester) become phase reference files the orchestrator-agent consults at role-switch boundaries. - Boss-context offload preserved; fresh-per-phase context given up (the property that drove or.1's nested-dispatch design is not buildable in Claude Code). Files touched: - skills/implement/agents/ailang-implement-orchestrator.md (frontmatter, Iron Law, Phase 2/3 rewrite, rationalisations, red flags) - skills/implement/SKILL.md (frontmatter, Iron Law, sub-status vocabulary note, cross-references) - skills/README.md (Conventions: no more named exception; agent roster: role-files recast as phase references) - docs/journals/INDEX.md (or.2 entry appended) - docs/journals/2026-05-11-iter-or.2.md (new — full rationale) - docs/roadmap.md (P1 tool-wiring item removed; resolved as categorically-not-fixable) - docs/WhatsNew.md (user-facing correction entry appended) No code changes; no bench impact; CLAUDE.md untouched (no stale references to fix there).
29 lines
2.9 KiB
Markdown
29 lines
2.9 KiB
Markdown
# What's New in AILang
|
|
|
|
A user-facing changelog. It runs in parallel to the technical
|
|
per-iter journals (`docs/journals/`) and is written for the
|
|
user-as-reader who did not watch the work happen — no crate names,
|
|
no iteration codes, no implementation mechanics.
|
|
|
|
Each entry corresponds to a done-state notification I sent to the
|
|
user when he was away. The text is identical to what arrived on his
|
|
phone. New entries are appended at the bottom.
|
|
|
|
What gets described here is **what changed in the project** — what
|
|
is now possible, fixed, or visible. For the internals, see
|
|
`docs/journals/INDEX.md` and the per-iter journal files it points at.
|
|
|
|
---
|
|
|
|
## 2026-05-11 — implement cycle no longer eats your context
|
|
|
|
The implement cycle no longer eats your context. From the next time we run it, the whole per-task review loop runs inside a dedicated subagent and sends back a short end-report — typing the chatter into someone else's window, so to speak. Each iter gets its own branch; merging or discarding is your call, and rollbacks become trivial. The 14k-line journal monolith is split into per-iter files with a small index for navigation; the old file is archived. Audit ran clean — no benchmark regressions; a small follow-up sweep fixed four spots where I'd missed stale references to the old layout.
|
|
|
|
## 2026-05-11 — plan-writing offload + first-run debrief
|
|
|
|
Plan-writing also offloads its read-heavy phase now — the part where I walk the codebase looking for the right files for a plan. Should feel lighter on the chat side from the next plan onwards. First real test of yesterday's per-task isolation refactor exposed a tool-wiring bug along the way: the Boss-context offload part works (you still see only an end-report at the end), but the per-task fresh-subagent isolation didn't actually fire — all tasks ended up running in one bigger context inside the orchestrator. Diagnosis is queued at the top of the next-up list. A small iter-branch setup wording bug got fixed in the same session. Audit clean, no benchmark regressions.
|
|
|
|
## 2026-05-11 — implement-cycle correction
|
|
|
|
Yesterday's note claimed the per-task review loop runs as separate fresh-context phases. That isn't possible — Claude Code does not allow a sub-helper to spawn further sub-helpers, full stop. The project's older "no helper calls another helper" convention turns out to have been a restatement of that platform rule all along, not a local design choice with a documented exception. So the architecture is corrected: the per-task loop still runs offloaded from your chat into one dedicated helper context (the part that does save your chat budget), but the review phases now run as sequential checkpoints inside that helper rather than as separate fresh contexts. The chat-side experience does not change. The fresh-per-phase isolation that the earlier note implied does not exist and never could; honestly scoped now.
|