iter boss: /boss skill — autonomous-mode discipline gated to user invocation

CLAUDE.md previously mixed universal facts (agent role boundaries,
commit discipline, design rationale, file roles, TDD-for-bugs) with
mode-specific autonomy rules (direction freedom, notifications,
WhatsNew procedure). Autonomous-by-default conflicted with the user's
intent that a fresh session should be collaborative-interactive unless
explicitly elevated.

Add `skills/boss/` containing only the three genuinely mode-specific
subsections — Direction freedom, Notifications, Done-state notifications:
WhatsNew.md. Trim CLAUDE.md from 343 to 243 lines; extend the skill-
system pointer paragraph with a one-sentence /boss gate. Universal
orchestrator discipline stays in CLAUDE.md because it applies whether
/boss is active or not.

Two cross-references that named the moved subsections by sub-heading
are repointed: skills/implement/SKILL.md and the implement-orchestrator
agent's standing reading list. The other ~11 agent-file references to
"orchestrator framing" still resolve correctly because that framing
stays in CLAUDE.md.

skills/README.md skill table extended with a `boss` row (now eight
skills); pipeline-diagram caption notes /boss wraps the pipeline.
.claude/skills/boss symlink follows the existing relative-path
convention.
This commit is contained in:
2026-05-12 10:14:33 +02:00
parent a99a39b65e
commit e953b137eb
9 changed files with 365 additions and 112 deletions
+1
View File
@@ -0,0 +1 @@
../../skills/boss
+6 -106
View File
@@ -48,6 +48,12 @@ are sharper tools, not a replacement for orchestrator judgement.
Specs go to `docs/specs/<milestone>.md`, plans to
`docs/plans/<iteration>.md`.
Autonomous orchestrator mode — picking the next iter from
`docs/roadmap.md` and looping until done-state — is gated to the
user-invoked `/boss` skill (`skills/boss/SKILL.md`). Outside
`/boss`, the default is interactive collaboration: the user asks,
Claude responds, Claude stops.
## My role: orchestrator
I am the **orchestrator** of this project, not the implementer. The
@@ -167,112 +173,6 @@ appeal does not count; neither does human ergonomics. Full criterion
lives in `docs/DESIGN.md` ("Feature-acceptance criterion") and is
applied as a gate by `skills/brainstorm/SKILL.md` during spec writing.
### Direction freedom
I have authority to choose the next iteration, refactor, or feature
without asking. Wrong calls are contained, not recovered: only the
Boss commits, and only when the working-tree state is consistent —
so bad work doesn't reach main in the first place. If a dispatched
agent's output is wrong, the remedy is `git checkout -- <paths>` on
the working tree (Boss-side), never a rewind of main HEAD. main is
forward-only.
The cost of asking "what should I do next" — context-switch for
the user, latency on my side — exceeds the expected cost of
discarding a bad working-tree state. So when the queue is non-empty
and the path is clear, just pick and proceed.
Bounce back to the user only when:
- A queued option requires a real design judgement I have not
made myself (genuine architectural fork, multiple substantive
options none of which is clearly default).
- I have hit something genuinely unexpected that changes the
project's direction (a fundamental design flaw, an external
dependency failure, a discovered invariant violation).
- The user has explicitly asked for a checkpoint.
A summary of what shipped is fine and welcome — but in
autonomous mode, follow it with the next dispatch, not a
question.
### Notifications
Two states call for a notify; nothing else does:
1. **Done-state.** The autonomous queue is empty — the milestone
closed, the audit ratified, and no further iteration is obvious
from the spec or the roadmap (or, in a user-scoped session,
every objective the user named has been completed and there is
nothing else queued). A finished sub-goal inside a still-open
milestone is *not* done-state. If the next iteration is obvious
and substantive, keep dispatching — "test suite landed", "audit
clean before close-fixes", "iter N done with N+1 in scope" are
internal progress, not notify events.
2. **Problem-state.** A bounce-back trigger per §"Direction
freedom" — design fork I cannot resolve, an unexpected
invariant violation, an external dependency failure, or the
user has explicitly asked for a checkpoint.
A user-specified stop trigger ("until X is done, then notify")
counts as done-state only if X is truly the last actionable item
in the autonomous queue at that moment. If hitting X still leaves
an open milestone with obvious follow-up iters, the right action
is to continue past X without notifying — the user's intent
behind the trigger is "wake me when there's nothing else to do",
not "wake me at this specific checkpoint".
Mid-flow progress notifications burn the user's attention without
giving him a decision to make. When in doubt, continue.
~/.claude/notify.sh "Text"
When notifying, the message body should be the actionable
summary: what I need from them, in one short line. Skip the
"hi, I" framing — just the gist. The user will see it on phone
and likely respond by returning to the session.
By default, I act autonomous. The notification is the
exception, not the rule.
### Done-state notifications: WhatsNew.md
When the trigger is **done-state** (autonomous work has wrapped
up, nothing left to do for the user to react to), the procedure
is two-step and the two outputs share one text:
1. Append a new entry to `docs/WhatsNew.md` (newest at the bottom,
append-only — same chronological convention as the per-iter journals).
2. Send the same text via `~/.claude/notify.sh`.
The text must be written for the **user-as-reader who did not
watch me work**:
- **Language: English.** Same as the rest of the repo; both
WhatsNew.md and the Notify push are permanent enough to be
treated as project content.
- **No technical internals.** Don't name crates, function
identifiers, type names, agent names, or iteration codes
(`ct.1.7`, `18a`, etc.). Describe the *change in the project*
— what is now possible, fixed, or visible — not the mechanics.
- **No assumption of context.** The user is on his phone and has
not seen this session. Lead with the result; the *why* is fine
if it fits in one short clause.
- **Length: Telegram-pragmatic.** No hard cap, but a few sentences,
not an essay. If the work covered multiple unrelated things,
bullet them.
- **Tone: factual, not performative.** No celebration, no "I have
successfully…". Just "X works now, Y is fixed, Z is new".
**Bounce-backs** (asking the user for a design judgement or
checkpoint) are *not* done-state and do **not** produce a
WhatsNew entry — they stay as Notify-only.
Format of a WhatsNew entry: a `## YYYY-MM-DD — <short title>`
heading followed by the message body. The body must match the
Notify text verbatim.
## Bug fixes — TDD, always
Bug fixes are RED-first, autonomous, no orchestrator gate. See
@@ -0,0 +1,12 @@
{
"iter_id": "boss",
"date": "2026-05-12",
"mode": "standard",
"outcome": "DONE",
"tasks_total": 6,
"tasks_completed": 6,
"reloops_per_task": { "1": 0, "2": 0, "3": 0, "4": 0, "5": 0, "6": 0 },
"review_loops_spec": 0,
"review_loops_quality": 0,
"blocked_reason": null
}
+69
View File
@@ -0,0 +1,69 @@
# 2026-05-12 — iter boss: `/boss` skill — autonomous-mode discipline gated to user invocation
## What shipped
- New skill at `skills/boss/SKILL.md`. Consolidates the three
mode-specific subsections that previously lived in `CLAUDE.md`:
Direction freedom, Notifications, Done-state notifications:
WhatsNew.md.
- `CLAUDE.md` refactored. Three subsections removed (~95 lines).
Skill-system pointer paragraph extended with one sentence
naming `/boss` as the autonomous-mode gate.
- `.claude/skills/boss` symlink created (`../../skills/boss`),
matching the discoverability convention codified in
`skills/README.md`.
- `skills/README.md` updated: skill table extended with a `boss`
row, heading shifted from "seven skills" to "eight skills",
pipeline diagram captioned with a `/boss`-wraps note.
- Two cross-references repointed: `skills/implement/SKILL.md`
and `skills/implement/agents/ailang-implement-orchestrator.md`
both named the "Done-state notifications" subsection of
CLAUDE.md by sub-heading; both now point at `skills/boss/SKILL.md`.
## Why
CLAUDE.md is always loaded into context. Mixing universal facts
(agent role boundaries, commit discipline, design rationale, file
roles, TDD-for-bugs) with mode-specific autonomy rules (direction
freedom, notifications, WhatsNew procedure) was OK while
autonomous-by-default was the operating assumption — but it
conflicted with the user's intent that a fresh session should be
collaborative-interactive unless explicitly elevated.
The mode-switch needs to be explicit. `/boss` is the explicit
elevation, and only the three genuinely mode-specific subsections
move with it. Universal orchestrator discipline stays in CLAUDE.md
because it applies whether `/boss` is active or not: agents need
to know role boundaries when dispatched in any session, design
discipline applies whenever a design choice is made (including a
`brainstorm` invoked outside `/boss`), and the orchestrator can
edit skills at any time.
## Decisions
- **Tightened the move-list from seven items to three.** The
initial spec proposed moving "My role: orchestrator", "Authority
over skills/", "Design rationale ≠ effort", "Feature acceptance",
"When NOT to delegate", plus the three mode-specific subsections.
Planner recon surfaced that the first five are universal, not
mode-specific — they apply in any session. The move-list was
narrowed to the three genuinely mode-specific items. The
brainstorm Step 7.5 grounding-check was re-dispatched on the
edited spec and returned PASS.
- **No private agents under `skills/boss/agents/`.** `/boss` is an
orchestration layer, not a worker. It dispatches the existing
skills (and their agents). The skill table reflects this by
listing `boss` as a meta-skill without an agent-roster row.
- **Cross-reference repoints kept tight.** Recon found 11+
agent/skill files that reference "CLAUDE.md — orchestrator
framing". Most of those resolve correctly to the universal
content that stays in CLAUDE.md. Only two files name the moved
subsections by sub-heading (`implement/SKILL.md` and the
orchestrator-agent's standing reading list); only those two
were repointed.
## Status
Iter closed. No bench regressions to check (markdown-only change,
no compiler touch). No audit needed (single-iter milestone, no
DESIGN.md change). Acceptance criteria from the spec all met.
+1
View File
@@ -19,3 +19,4 @@
- 2026-05-12 — iter rt.1: roundtrip invariant audit tests (3 new tests + ailx-pair dynamic); all PASS first-shot, no gaps surfaced → 2026-05-12-iter-rt.1.md
- 2026-05-12 — iter rt.2: DESIGN.md anchor for Roundtrip Invariant (top-level section + Decision 6 Constraint 2 upward cross-reference) → 2026-05-12-iter-rt.2.md
- 2026-05-12 — audit-rt: milestone close (Roundtrip Invariant) — architect drift fixed in rt.tidy (3 items: Direction-2 5th test, roadmap P1 removed, wording sync); bench all-green → 2026-05-12-audit-rt.md
- 2026-05-12 — iter boss: `/boss` skill — autonomous-mode discipline (Direction freedom, Notifications, WhatsNew procedure) extracted from CLAUDE.md into a user-invoked skill → 2026-05-12-iter-boss.md
+6 -1
View File
@@ -10,7 +10,7 @@ The system was bootstrapped on 2026-05-09. See
`docs/specs/2026-05-09-skill-system.md` for the design and
`docs/journal-archive.md` ("Skill system live") for the rationale.
## The seven skills
## The eight skills
| Skill | Trigger | Output | Mandatory? |
|-------|---------|--------|------------|
@@ -21,9 +21,14 @@ The system was bootstrapped on 2026-05-09. See
| [`docwriter`](docwriter/SKILL.md) | API surface stabilized; rustdoc lag suspected | rustdoc updates in `///` and `//!`, uncommitted in the working tree | No — Boss-dispatched only |
| [`fieldtest`](fieldtest/SKILL.md) | Boss-dispatched post-audit field test on a milestone that touched user-visible surface | 2-4 `.ailx` example fixtures + `docs/specs/<date>-fieldtest-<milestone>.md`, all uncommitted in the working tree | No — Boss-dispatched only |
| [`debug`](debug/SKILL.md) | Bug encountered (failing test, segfault, wrong stdout) | RED-test in the working tree (uncommitted) + cause analysis | **Mandatory RED-first** for any bug |
| [`boss`](boss/SKILL.md) | User-invoked only (`/boss`) | Autonomous orchestration session — dispatches existing skills until done-state or bounce-back | No — user-gated |
## Pipeline
> Inside a `/boss` session the orchestrator walks this pipeline
> autonomously. Outside `/boss`, the pipeline runs only when the
> user explicitly invokes a skill.
```
[new milestone] [bug observed]
| |
+263
View File
@@ -0,0 +1,263 @@
---
name: boss
description: User-invoked only (typed as `/boss`). Activates autonomous orchestrator mode — Claude picks the next iteration from `docs/roadmap.md`, dispatches the appropriate skill, and continues until done-state or genuine bounce-back. Outside `/boss`, the default is interactive collaboration with the user, not autonomous queue execution.
---
# boss — autonomous orchestrator mode
> **Violating the letter of these rules is violating the spirit.**
## Overview
Autonomous orchestrator mode is the way Brummel-as-LLM moves the
AILang project forward when the user is away. Inside this mode,
Claude reads `docs/roadmap.md`, picks the top item, dispatches the
right skill, integrates the result, and continues to the next item
— until the queue is empty (done-state) or a real design fork
forces a bounce-back to the user.
Outside this mode, the default is interactive collaboration: the
user asks; Claude answers or executes the specific request; Claude
stops. There is no autonomous picking from the roadmap, no
multi-iter sequence chained off a single user message.
This skill exists to make the mode-switch explicit. Without it,
autonomous behaviour and interactive behaviour both fight for
"default" in every fresh session, with the failure mode of an
interactive session silently turning into an autonomous one (or
vice versa).
## When to Use / Skipping
Triggers:
- The user explicitly types `/boss` in chat.
Never auto-invoked. The using-superpowers rule that the harness
applies — "if a skill might apply, invoke it" — does NOT apply to
this skill. `/boss` is a deliberate mode switch performed by the
user, not a heuristic that Claude resolves on its own. A session
that wanders into autonomous behaviour without `/boss` is a
failure mode this skill exists to prevent.
May be exited mid-session by:
- The user signalling "stop", "pause", or "back to chat".
- A bounce-back trigger firing (see "Direction freedom" below).
- Done-state (the queue is empty and the done-state procedure has
fired).
A fresh session always starts in interactive mode; the user
re-enters by typing `/boss` again.
## The Iron Law
```
NO AUTONOMOUS DISPATCH OUTSIDE A `/boss` SESSION.
DONE-STATE = QUEUE EMPTY, NOT SUB-GOAL COMPLETE.
BOUNCE-BACK ONLY ON THE THREE NAMED TRIGGERS.
WHATSNEW.MD + NOTIFY ARE ONE TEXT, NOT TWO.
```
Universal rules (only-Boss-commits, main-is-sacrosanct, language,
TDD-for-bugs) come from `CLAUDE.md` and apply regardless of
mode. This skill does not restate them.
## The Process
### Step 1 — Read the queue
`docs/roadmap.md` is the priority-ordered forward queue. Read it
first. The top item is the candidate for the next dispatch.
If the queue is empty: skip to Step 5 (done-state). Do not invent
work; an empty queue is a real signal that the milestone closed
and there is nothing left.
### Step 2 — Pick the top item per direction-freedom rules
See "Direction freedom" below. The rules permit picking
autonomously when the path is clear. If the top item is a real
fork (two substantive options with no clear default), bounce back
to the user.
### Step 3 — Dispatch the appropriate downstream skill
Pipeline (from `skills/README.md`):
```
brainstorm → planner → implement → audit → fieldtest
+
debug (bug-triggered)
+
docwriter (post-stability)
```
Match the chosen item to the right entry skill. If a milestone is
new and no spec exists: `brainstorm`. If a spec exists but no plan:
`planner`. If a plan exists: `implement`. If a milestone is closing:
`audit`. If the iter is a bugfix: `debug`. Read each skill's
SKILL.md trigger section if unsure.
If the working tree is mid-flight (uncommitted changes left over
from a previous session): inspect first, then resume the right
step. Do not restart from scratch.
### Step 4 — Loop, deciding on each agent output
For each dispatched skill's result, decide:
- **Continue** — agent output is good, next item is obvious, the
queue is non-empty. Most common case.
- **Bounce-back** — one of the three named triggers fired (see
"Notifications" below). Send a notify, stop the loop.
- **Done-state** — queue is empty AND the milestone is fully
ratified (audit clean, fieldtest clean if applicable). Go to
Step 5.
### Step 5 — On done-state: WhatsNew.md + Notify
Run the procedure named in "Done-state notifications: WhatsNew.md"
below. The two outputs share one verbatim text.
## Direction freedom
I have authority to choose the next iteration, refactor, or feature
without asking. Wrong calls are contained, not recovered: only the
Boss commits, and only when the working-tree state is consistent —
so bad work doesn't reach main in the first place. If a dispatched
agent's output is wrong, the remedy is `git checkout -- <paths>` on
the working tree (Boss-side), never a rewind of main HEAD. main is
forward-only.
The cost of asking "what should I do next" — context-switch for
the user, latency on my side — exceeds the expected cost of
discarding a bad working-tree state. So when the queue is non-empty
and the path is clear, just pick and proceed.
Bounce back to the user only when:
- A queued option requires a real design judgement I have not
made myself (genuine architectural fork, multiple substantive
options none of which is clearly default).
- I have hit something genuinely unexpected that changes the
project's direction (a fundamental design flaw, an external
dependency failure, a discovered invariant violation).
- The user has explicitly asked for a checkpoint.
A summary of what shipped is fine and welcome — but in
autonomous mode, follow it with the next dispatch, not a
question.
## Notifications
Two states call for a notify; nothing else does:
1. **Done-state.** The autonomous queue is empty — the milestone
closed, the audit ratified, and no further iteration is obvious
from the spec or the roadmap (or, in a user-scoped session,
every objective the user named has been completed and there is
nothing else queued). A finished sub-goal inside a still-open
milestone is *not* done-state. If the next iteration is obvious
and substantive, keep dispatching — "test suite landed", "audit
clean before close-fixes", "iter N done with N+1 in scope" are
internal progress, not notify events.
2. **Problem-state.** A bounce-back trigger per §"Direction
freedom" — design fork I cannot resolve, an unexpected
invariant violation, an external dependency failure, or the
user has explicitly asked for a checkpoint.
A user-specified stop trigger ("until X is done, then notify")
counts as done-state only if X is truly the last actionable item
in the autonomous queue at that moment. If hitting X still leaves
an open milestone with obvious follow-up iters, the right action
is to continue past X without notifying — the user's intent
behind the trigger is "wake me when there's nothing else to do",
not "wake me at this specific checkpoint".
Mid-flow progress notifications burn the user's attention without
giving him a decision to make. When in doubt, continue.
~/.claude/notify.sh "Text"
When notifying, the message body should be the actionable
summary: what I need from them, in one short line. Skip the
"hi, I" framing — just the gist. The user will see it on phone
and likely respond by returning to the session.
By default, I act autonomous. The notification is the
exception, not the rule.
## Done-state notifications: WhatsNew.md
When the trigger is **done-state** (autonomous work has wrapped
up, nothing left to do for the user to react to), the procedure
is two-step and the two outputs share one text:
1. Append a new entry to `docs/WhatsNew.md` (newest at the bottom,
append-only — same chronological convention as the per-iter journals).
2. Send the same text via `~/.claude/notify.sh`.
The text must be written for the **user-as-reader who did not
watch me work**:
- **Language: English.** Same as the rest of the repo; both
WhatsNew.md and the Notify push are permanent enough to be
treated as project content.
- **No technical internals.** Don't name crates, function
identifiers, type names, agent names, or iteration codes
(`ct.1.7`, `18a`, etc.). Describe the *change in the project*
— what is now possible, fixed, or visible — not the mechanics.
- **No assumption of context.** The user is on his phone and has
not seen this session. Lead with the result; the *why* is fine
if it fits in one short clause.
- **Length: Telegram-pragmatic.** No hard cap, but a few sentences,
not an essay. If the work covered multiple unrelated things,
bullet them.
- **Tone: factual, not performative.** No celebration, no "I have
successfully…". Just "X works now, Y is fixed, Z is new".
**Bounce-backs** (asking the user for a design judgement or
checkpoint) are *not* done-state and do **not** produce a
WhatsNew entry — they stay as Notify-only.
Format of a WhatsNew entry: a `## YYYY-MM-DD — <short title>`
heading followed by the message body. The body must match the
Notify text verbatim.
## Common Rationalisations
| Excuse | Reality |
|--------|---------|
| "Iter just landed cleanly, let me notify so user can see" | Mid-flow notifications burn user attention without giving them a decision to make. Iter-clean is internal progress, not a notify event. |
| "Sub-goal of the milestone is done, that counts as done-state" | Done-state is queue-empty, not sub-goal-complete. If the next iter is obvious and substantive, keep dispatching. |
| "Top of roadmap has two equally good options, I'll just pick one" | If both are substantive and the choice is non-obvious, that's a real fork — bounce back. The cost of a wrong pick (working-tree discard) is higher than the cost of one short ping to the user. |
| "User said 'until X then notify', X just completed, notify now" | Check whether X was actually the last actionable thing. If hitting X leaves an open milestone with obvious follow-up, the user's intent was "wake me when there's nothing else to do" — continue. |
| "Done-state hit, send notify, append WhatsNew later" | WhatsNew + notify are one text. Append first (atomic), send second. Splitting them is how the WhatsNew gets forgotten. |
| "WhatsNew can name the iter code — user will figure out what it means" | No. WhatsNew is for the user-as-reader who did not watch me work. Iter codes, crate names, function identifiers, type names — all internal, all banned. |
## Red Flags — STOP
- About to send a notify on an internal-progress event (test suite landed, audit clean before close-fixes, iter N done with N+1 in scope).
- About to dispatch the next skill without checking that the previous output was actually good.
- About to autonomously dispatch in a session where the user did NOT type `/boss`.
- About to write a WhatsNew entry that names a crate, an iter code, or an agent.
- About to bounce back at every iter boundary "just to be safe" — that's reactive deference, not direction freedom.
- About to send the WhatsNew text and the Notify text with different wording — they must be one verbatim text.
## Cross-references
- **Skill system index:** `skills/README.md` — skill table, agent
roster, pipeline diagram.
- **Universal rules:** `CLAUDE.md` — agent role boundaries, commit
discipline, design rationale, authority over skills, feature
acceptance, TDD-for-bugs, file-role roster.
- **Queue:** `docs/roadmap.md` — priority-ordered forward queue
consumed by Step 1.
- **Canonical feature-acceptance criterion:** `docs/DESIGN.md`
§"Feature-acceptance criterion" — the criterion that gates new
features; applied at spec time by `skills/brainstorm`.
- **Downstream skills dispatched:** `skills/brainstorm`,
`skills/planner`, `skills/implement`, `skills/audit`,
`skills/fieldtest`, `skills/debug`, `skills/docwriter`.
+3 -2
View File
@@ -141,8 +141,9 @@ unstaged changes. Nothing is committed yet.
the same commit (or a follow-up commit):
`- YYYY-MM-DD — iter <iter_id>: <one-line title> → <YYYY-MM-DD>-iter-<iter_id>.md`
6. If trigger is done-state and the user is away, write a
`docs/WhatsNew.md` entry + run `notify.sh` per CLAUDE.md's
"Done-state notifications" subsection.
`docs/WhatsNew.md` entry + run `notify.sh` per
`skills/boss/SKILL.md` "Done-state notifications: WhatsNew.md"
subsection.
### Step 4 — Boss handling (on BLOCKED)
@@ -28,9 +28,10 @@ context and dies with it.
Read these before doing anything else, in this order:
1. `CLAUDE.md` — orchestrator framing, in particular the
"Done-state notifications: WhatsNew.md" subsection (WhatsNew is
Boss-side, NOT this agent's job).
1. `CLAUDE.md` — orchestrator framing (agent role boundaries,
commit discipline). Note: the "Done-state notifications:
WhatsNew.md" procedure has moved to `skills/boss/SKILL.md`;
WhatsNew is Boss-side regardless of mode, NOT this agent's job.
2. `docs/DESIGN.md` — invariants any iter must respect.
3. `docs/journals/INDEX.md` plus the last 13 per-iter journal
files it points at — recent state of the project.