mirror of
https://github.com/garrytan/gstack.git
synced 2026-11-17 05:41:50 +01:00
Merge origin/main into gsta-167-review-packets; v1.91.326.0 stamp
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
This commit is contained in:
commit
60a23de4bf
17 files changed
+640
-92
No files matched your search
+40
-9
@@ -50,7 +50,7 @@ With a cap set and no readable meter, start refuses
|
||||
|
||||
## The BUDGET line
|
||||
|
||||
BUDGET: used=<n> reserved=<n> unknown=<n> admission_cap=<n|none> tail=<n> unit=<u> meter=<m> epoch=<n> call=<n|unknown> next=<id> projected=<n> calibration=step:<f|prior>,dispatch:<f|prior> verdict=continue|delegate|stop
|
||||
BUDGET: used=<n> reserved=<n> unknown=<n> admission_cap=<n|none> tail=<n> unit=<u> meter=<m> epoch=<n> call=<n|unknown> call_floor=<n> above_floor=<n|unknown> next=<id> projected=<n> calibration=step:<f|prior>,dispatch:<f|prior> verdict=continue|delegate|stop
|
||||
|
||||
| Field | Meaning |
|
||||
|---|---|
|
||||
@@ -61,6 +61,7 @@ With a cap set and no readable meter, start refuses
|
||||
| `tail` | Tail reserve (scaled by the step factor once measured) plus stop overhead, held back from every admission |
|
||||
| `unit`, `meter`, `epoch` | The meter contract: `capy-tokens` (root counter), `claude-usage-tokens` (journal usage), `estimate-tokens`; a meter change opens a new epoch |
|
||||
| `call` | The per-call charge: two Capy readings in consecutive calls, or the last Claude Code call's usage |
|
||||
| `call_floor`, `above_floor` | The host's per-call floor (see Delegation) and the charge above it, max(0, call − floor); the delegate rule compares `above_floor` |
|
||||
| `next`, `projected` | What the check was for and its projection: the prior times the run's factor for that kind |
|
||||
| `calibration` | The run's step and dispatch factors (see Calibration), or `prior` before the first measurement |
|
||||
| `verdict` | `continue` (exit 0, prints `RESERVATION:` under a cap), `delegate` (exit 0), `stop` (exit 3) |
|
||||
@@ -79,6 +80,11 @@ gstack-ship-budget check --run <id> --step 9 --next dispatch:specialist,red-team
|
||||
gstack-review-log --start review --reservation <id>
|
||||
```
|
||||
|
||||
Each reader is priced at max(its projection, the run's measured mean for that role over settled
|
||||
dispatches). When the whole batch does not fit, `check` admits the largest prefix that does and
|
||||
prints `ADMIT: k of n (remaining=<roles>)`: dispatch those k, then re-check the remaining roles
|
||||
once they settle. Only a batch whose first reader does not fit stops, and `close` flags remaining
|
||||
roles no later check requested (`DROPPED:`, status=incomplete until the owner accepts).
|
||||
Roles are review-depth's: `specialist`, `red-team`, `adversarial-native`, `outside-voice`,
|
||||
`fix-check`. /ship passes the `BUDGET_NEXT: dispatch:<roles>` value that the batch's
|
||||
`gstack-review-packet build` prints, one role per packet: a Step 7/8 audit reserves as
|
||||
@@ -95,27 +101,50 @@ BUDGET_CHECK_SKIPPED line and the run ends status=incomplete for the owner to ac
|
||||
|
||||
## Delegation
|
||||
|
||||
At start, a per-call charge of 100k or more delegates the whole /ship. Before Step 7 and before
|
||||
Step 12, 150k or more (or an unknown charge) delegates the next segment (7–11.5, 12–22):
|
||||
The delegate rule measures context above the host's per-call floor: what one call costs in a
|
||||
fresh session with the ship skill loaded, before any /ship work. The floor is `call_floor` per
|
||||
unit in `lib/ship-budget-calibration.json` (150k on Capy, where a fresh thread's calls cost
|
||||
150–230k; 40k on Claude Code; 0 for the estimate meter, whose charge is always unknown), or
|
||||
`start --call-floor <n>`, which the ledger records and every child inherits. At start, 100k or
|
||||
more above the floor delegates the whole /ship. Before Step 7 and before Step 12, 150k or more
|
||||
above the floor (or an unknown charge) delegates the next segment (7–11.5, 12–22). The BUDGET
|
||||
line prints `call`, `call_floor` and `above_floor`, so each verdict can be checked:
|
||||
|
||||
1. `gstack-ship-budget handoff --run <id> --from 12 --to 22 --base origin/<base> --reading <n>` writes a
|
||||
brief (mode 0600, under the run directory) with the pins (HEAD, merge base, working tree,
|
||||
reviewed tree), the record with its six headings and start tokens, the decisions in scope
|
||||
with the owner's words, the decision trail, the budget and the return contract. It reserves
|
||||
the child's sub-cap: min(cap − used − reserved − unknown − the calibrated tail reserve and stop
|
||||
overhead, the segment estimate), or `--sub-cap`.
|
||||
overhead, the segment estimate), or `--sub-cap`. A child handing off reserves its whole
|
||||
remaining sub-cap instead (cap − used − reserved − unknown − its stop overhead).
|
||||
2. The child is a Capy subagent on the shared machine, or on Claude Code
|
||||
`gstack-ship-budget launch --in <brief>`: a background `claude -p` (resolved through lib/claude-bin.ts)
|
||||
with explicit permission flags and the return file named.
|
||||
3. The child runs `verify-handoff` first; a moved pin refuses
|
||||
([HANDOFF_PIN_MOVED](troubleshooting.md#handoff-pin-moved)). It admits against its sub-cap
|
||||
with child_used = reading − reading at handoff, is the only writer of the record, and never
|
||||
delegates again: a delegate rule stops it. It ends with `return` (at most 40 lines).
|
||||
with child_used = reading − reading at handoff and is the only writer of the record. The
|
||||
context rule never stops a child. When it fires at one of the child's boundaries:
|
||||
- once the child has completed a step since `verify-handoff`, `verdict=delegate` hands its
|
||||
remaining steps (the next step through the end of its segment) to a fresh grandchild
|
||||
through these same four steps. The chain depth (1 for a child, 2 for a grandchild, at most
|
||||
3) is recorded in the brief, the handoff and the new ledger;
|
||||
- before its first completed step, at depth 3, or when the next step is past its segment,
|
||||
`verdict=continue` with one `CONTEXT_HIGH:` line (call, floor, threshold, depth, reason).
|
||||
|
||||
The admission cap still stops a child (`reason=budget_exhausted`). It reconciles any
|
||||
grandchild, then ends with `return` (at most 40 lines). `return` reports `status=incomplete`
|
||||
while one of the child's own handoffs is unreconciled (`UNRECONCILED:`) or an ADMIT remainder
|
||||
was never re-checked (`DROPPED:`), so a delegated segment cannot drop readers or leave a
|
||||
grandchild writing after the parent moves on.
|
||||
4. `gstack-ship-budget reconcile --run <id> --return <file> --reading <n>`; idempotent per handoff hash,
|
||||
so a duplicate return charges nothing twice. A refused, incomplete or interrupted child
|
||||
exits 3: regenerate the handoff or stop.
|
||||
exits 3: regenerate the handoff or stop. Up a chain each token is charged once: the Capy
|
||||
counter already includes the whole subtree, a session `launch` starts joins the Claude Code
|
||||
meter of every ancestor, and under `meter=estimate` a child's returned `used` carries its
|
||||
grandchild's.
|
||||
|
||||
`--delegate never` turns delegate into stop for hosts without children.
|
||||
`--delegate never` turns delegate into stop for a parent on a host without children; a child
|
||||
continues with a `CONTEXT_HIGH:` note instead.
|
||||
|
||||
## Stop and resume
|
||||
|
||||
@@ -147,7 +176,9 @@ largest observation times 1.5, rounded up to 10,000 tokens. The leave-one-out te
|
||||
(test/ship-budget-calibration.test.ts) covers the per-call constant only, the one observed
|
||||
across three runs; per-dispatch has two observations and stop overhead one. Only the Capy
|
||||
meter has observations; the Claude Code and estimate units are seeded from it until a
|
||||
Claude-metered /ship completes.
|
||||
Claude-metered /ship completes. Each unit's `call_floor` is calibrated separately, with its
|
||||
evidence in `call_floor_why`; a floor is the largest 10,000 multiple at or below its smallest
|
||||
observed fresh call.
|
||||
|
||||
These constants are the prior. Each settled step or dispatch is measured on the next fresh
|
||||
reading (a step from its check's reading to the next one; dispatches settled together as one
|
||||
|
||||
@@ -1832,9 +1832,10 @@ GSTACK_RESULT lines.
|
||||
<a id="budget-cap-reached"></a>
|
||||
### `gstack-ship-budget: ... (BUDGET_CAP_REACHED)`
|
||||
|
||||
**Meaning.** The next step or dispatch batch would push used + open reservations + unknown
|
||||
charges + its projection + the tail and stop reserve past the admission cap, so it did not
|
||||
start. Exit 3. `gstack-review-log --start` prints the same code (exit 1) for a dispatch
|
||||
**Meaning.** The next step, or the first reader of a dispatch batch, would push used + open
|
||||
reservations + unknown charges + its projection + the tail and stop reserve past the admission
|
||||
cap, so it did not start. Exit 3. A batch that partly fits is not refused: `check` admits the
|
||||
largest prefix that fits and prints `ADMIT: k of n (remaining=<roles>)`. `gstack-review-log --start` prints the same code (exit 1) for a dispatch
|
||||
after the run stopped. The cap admits work; it cannot stop a turn already running, so the
|
||||
overshoot is bounded by the misprojection of work admitted before the stop.
|
||||
|
||||
|
||||
Reference in new issue
Block a user