diff --git a/README.md b/README.md index de392e2..7e613cc 100644 --- a/README.md +++ b/README.md @@ -271,7 +271,7 @@ antigravity already has a great `/plan` which works great with pstack. but perso type [`/automate-me`](./skills/automate-me/SKILL.md). it mines your recent transcripts, drafts a `-mode` skill from how you've actually worked, and routes through pstack underneath. you keep pstack as the base and end up with your own routing skill alongside `poteto-mode`. -models are configurable too. type [`/setup-pstack`](./skills/setup-pstack/SKILL.md). it detects the models you have access to and writes a small always-applied rule mapping each role (code, judgment, the review panels) to a model. every skill reads it and falls back to sensible defaults when the rule is absent, so you override only what you want. +models are configurable too. type [`/setup-pstack`](./skills/setup-pstack/SKILL.md). it writes a small always-applied rule mapping each role (code, judgment, the review panels) to a tier (`pro`, `flash`, `flash_lite`, or `inherit`). every skill reads it and falls back to sensible defaults when the rule is absent, so you override only what you want. when a default changes, a rule written before the change still pins the old default. delete those role lines, or delete the file, then run `/setup-pstack` again. a rerun keeps any role whose model differs from the default. diff --git a/skills/setup-pstack/SKILL.md b/skills/setup-pstack/SKILL.md index da961f8..0f52d69 100644 --- a/skills/setup-pstack/SKILL.md +++ b/skills/setup-pstack/SKILL.md @@ -26,7 +26,7 @@ The default role-to-model mapping is the rule shape shown in step 5 below. If `~ - `medium - flash reasoning` - `small - flash_lite reasoning` -**(b) Apply it.** Build the working table from the skill defaults, and on a re-run keep any role you changed by tier, list, or alias (`inherit`, `inherit-parent`, `auto`). `unlimited` upgrades `flash` roles to `pro`. `large` keeps the skill defaults (`pro` for judgment, prose, and hardest tasks; `flash` for fast code, explorer, investigator, and worker roles). `medium` maps code and explorer roles to `flash` and judgment roles to `pro` or `flash`. `small` maps fast code and explorer roles to `flash_lite` and judgment roles to `flash`. `inherit`, `inherit-parent`, and `auto` do not change. +**(b) Apply it.** Build the working table from the skill defaults, and on a re-run keep any role you changed by tier, list, or alias (`inherit`, `inherit-parent`, `auto`). `unlimited` upgrades `flash` roles to `pro`. `large` keeps the skill defaults (`pro` for judgment, prose, and hardest tasks; `flash` for fast code, explorer, investigator, and worker roles). `medium` sets every role to `flash`. `small` sets every role to `flash_lite`. `inherit`, `inherit-parent`, and `auto` do not change. **(c) Show the roles and confirm.** Show every role with its model tier. Also list each line step 2 dropped. Ask whether to accept as-is or change specific roles, offering `pro`, `flash`, `flash_lite`, and `inherit` (which runs that role on the parent chat model and preserves its reasoning effort) as the options. Prefer `ask_question` over free text. For panel roles (arena runners, architect runners, interrogate reviewers) the value is a list, and one subagent runs per entry, alias entries included, so the list length sets the count. `arena cross-judge pool` is also a list, and Arena selects one value from it that differs from the parent's tier when possible. `swarm workers` is the default tier for every worker unless a race or comparison assigns another tier per arm.