cafleet: Skill for Claude Code

CAFleet Model List Refresh

Refresh the repository's model list — the per-backend model tables of skills/cafleet/reference/model-list.md — from the approved official sources, under explicit maintainer review. The repository file is the release source: a running Director reads only its deployed skill replica, so a refresh reaches Directors exclusively through the release/deployment transaction below.

Approved sources (exhaustive allowlist)

Fetch only these five official pages. Search results, third-party price sites, and social posts are never model-list authority. The first four are pricing and availability authority; the fifth is capability authority for the claude backend's context windows and Claude Code model-string syntax, and is never used to price a model.

Refresh procedure

1. Fetch all five approved sources. If any source cannot be fetched or parsed, stop: leave the model list unchanged, report the error, and let it become stale rather than fabricating values.
2. Extract pricing facts only: the currently effective standard input and output prices (USD per MTok) and availability/deprecation qualifiers. Do not copy long provider text into the repository, and do not present the pricing pages as benchmark evidence — they are pricing/availability sources only.
3. Refresh the claude and codex tables from their provider pricing pages, and refresh the opencode table with the dedicated procedure below — the Zen catalog behaves differently from the provider pages, so follow that procedure exactly. For codex, pricing and availability come from different pages: price each model from the OpenAI pricing page, but take Codex availability from the Codex models page (https://learn.chatgpt.com/docs/models.md) — the table carries only the models Codex currently offers in its model picker. A model absent there or demoted to a legacy model (reachable only via codex -m / config.toml) is removed from the table even when the pricing page still prices it; legacy models remain manual pass-through --model values.
4. Reapply the capability classes explicitly. The Class descriptions and the most-to-least-capable row ordering are reviewed maintainer judgment, not provider benchmark claims; changing a class or the ordering requires reviewed policy approval in the same pull request. Every row carries both prices from its approved source. Re-derive the Monitor and reviewer defaults table from the refreshed backend tables, and mirror its values into each backend section's {monitor_model} and {reviewer_model} rows (skills/cafleet/reference/coding-agent-overlays.md).
5. Re-verify every claude row's context window on every refresh, even when the prices are unchanged. Take each window from the Claude Code model configuration page's Extended context section and record it in the claude table's Context column. That page is also the authority for whether a row's --model value carries the [1m] suffix, which is a Claude Code model-string suffix stripped before the request reaches the provider — it is meaningless for the codex and opencode tables, whose values never carry it. Apply the suffix only where it changes the window a member actually gets:
   • A model the page lists as always running at 1M on the Anthropic API carries no suffix — the window is already 1M and the suffix is inert.
   • A model the page gives no 1M variant carries no suffix.
   • A model without 1M support carries no suffix; appending it there is a malformed value.
   • Where the page shows the window is opt-in for the reader's plan, the suffix belongs in the page's note rather than the row, so the Director applies it per spawn against the operator's plan instead of every spawn inheriting a credit-billed window.