Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 17 additions & 3 deletions AGENTS.md
Original file line number Diff line number Diff line change
Expand Up @@ -59,9 +59,23 @@ overridden. Never use `route-override`, `--est-pct`, `--min-headroom 0` or
wont-fit, inflight, headroom or preflight), when `aiuse --available --json`
lists it `usable_now: false`, or, for Argon, when `acp-run agy` fails on
quota. Short means move to the next model in the order, never force it.
Say in the PR body which model was short and why. As of 2026-10-31 astra
is short: Codex has 45% left, below the 60% `codex:frontier` estimate
floor, so Fable takes the first seat.
Say in the PR body which model was short and why. **Check the actual
dispatch, never infer availability from the config alone.** Run the
intended command with `--dry-run`, preserving its model, kind, effort,
brief, `-C` and output/task directory. Inspect `would refuse: class=...`
(or `would_refuse_class` with `--json`), not just the exit status:
dry-run returns 0 for predicted headroom, fit and concurrency refusals.
It skips preflight, so no predicted refusal is not proof of availability;
the real dispatch must still pass its checks. The `codex:frontier`
estimate uses comparable measured history: on 2026-11-01 the selected
machine-wide history for astra at high effort held 14 matching records,
and the scalar and weekly-window estimates were both 2 percentage points
(upper median). The configured 60% fallback floor applies when the
selected history has fewer than three measurements for the scalar or the
relevant quota window; three measurements overall do not release every
window's floor. On 2026-10-31 a lead treated the configured 60% fallback
as a permanent minimum and declared astra short at 45% remaining. That
comparison alone did not establish that astra was short.
- Design rulings (a `schema_version` bump, the privilege or enforcement
scope on a host, a deploy plan, the unit-writer or consent-diff design)
get two independent frontier reviews, astra and Fable. Argon stands in
Expand Down
6 changes: 4 additions & 2 deletions docs/architecture/schema-v4-reconciliation-2026-10-31.md
Original file line number Diff line number Diff line change
Expand Up @@ -5,8 +5,10 @@ It supersedes its three inputs, which stay frozen as written:
[the Opus 5.5 draft](schema-v4-opinion-opus-draft.md), the
[Fable 5.1 review](schema-v4-opinion-fable.md) and the
[Gemini 4 Argon review](schema-v4-opinion-argon.md). Argon took the seat that the
standing rule gives to astra, because astra was short (Codex had 45% left, below the
60% `codex:frontier` estimate floor). The standing rule is in `AGENTS.md` on PR
standing rule gives to astra. The lead called astra short by comparing Codex's 45%
remaining with the 60% `codex:frontier` fallback in the config. That comparison did not
establish it: the dispatch estimate uses measured history, and an astra review dispatched
on 2026-11-01 at 41% was admitted. Both reviews are still cross-model. The standing rule is in `AGENTS.md` on PR
https://github.com/frdminc/tendcf/pull/24 and was reviewed there. Where an input is
wrong, §8 says so; the inputs are not corrected in place. This document was itself
reviewed by Fable 5.1 (approve with changes); those changes are applied.
Expand Down
Loading