bermudis-pi-goodies

A bundle of small, frequently-used Pi extensions. One entry point, ten independent features.

Feature	Command / hook	What it does
copy-with-model	/copy-with-model	Copy last assistant message to the clipboard in a code fence tagged with the model name.
copy-trajectory	/copy-trajectory [thinking]	Copy the whole conversation (user + assistant text, tool calls stripped) to the clipboard; thinking also includes assistant thinking blocks.
name-with-ai	/name-with-ai [name]	Generate a short session name from the first user message (or set one manually).
zed	/z	Open Zed editor on the current working directory.
prefer-tools	hook (no command)	Nudge toward modern CLIs: rg over grep, fd over find, uv over bare python/pip/pytest/mypy.
keep-model-on-new	hook (no command)	Keep the active model when /new starts a fresh session instead of reverting to pi's saved default model.
model-thinking	hook + /levels	Per-model thinking levels stored in an extension sidecar (pi's /scoped-models rewrites enabledModels bare, wiping :level suffixes); /levels edits them, hooks apply them whenever a model becomes active.
kilo	provider	Access Kilo Gateway models via /login kilo or KILO_API_KEY.
provider-balance	footer (no command)	Show remaining Kilo or OpenRouter credits, z.ai token-plan quota, or OpenAI Codex quota on the right side of the working-directory footer line.
tps	hook (no command)	Notify tokens/sec and in/out/cache token usage at the end of each agent turn.

Install

After publishing the package to npm:

pi install npm:bermudis-pi-goodies@0.6.1

Remove any old bermudis-pi-goodies.ts symlink before reloading Pi. Each feature is independent — disabling one is a one-line edit in index.ts. Kilo's provider and its balance footer are bundled here.

Model selection and thinking

Pi's native scoped-models config would put per-model thinking levels in enabledModels entries like "zai/glm-5.3:high", but the /scoped-models screen (which maintains that list) rewrites it with bare model ids on every save, destroying any :level suffix. So this package stores levels in its own sidecar file:

~/.pi/agent/data/bermudis-pi-goodies/thinking-levels.json

The split of ownership is clean: /scoped-models owns which models are enabled and their Ctrl+P cycle order; /levels owns the per-model thinking level and can never be wiped by the other screen.

Use /levels to edit: ↑↓ moves between scoped models, ←→ cycles each model's level through the ladder it actually supports plus inherit (which leaves pi's global default in charge), Enter saves atomically and applies to the active model immediately, Esc cancels.

Hooks apply the stored level whenever a model becomes active: full-picker selection, Ctrl+P cycling, startup, and /new (where this package carries the previous session's active model forward). A native scoped level for the session — via --models "x:high" or a hand-suffixed enabledModels entry — always wins; the sidecar only fills in where pi itself has no level. Explicit --thinking or --model ...:<level> still wins for that launched session — a model whose registered id genuinely ends in :<level> is indistinguishable from that shorthand without the registry, so it opts out of its stored level for that startup; switch or cycle to re-apply. Resumed and forked sessions keep the level restored by pi.

One gap is inherent to Pi: it emits no model_select when you pick the model that is already active, so re-selecting it in the full picker leaves a manual thinking level in place — switch models (or cycle) to snap back. Saving the /scoped-models screen in pi 0.84.2 writes bare model ids, which is now harmless: bare ids are exactly what this design expects.

Provider and balance details

Kilo registration is network-free: it starts with kilo-auto/free, restores an authenticated catalog from Pi's model store, and normally revalidates that catalog no more than every four hours. Balance and quota requests run in the background so they never delay session startup, model selection, or post-run input readiness. The footer also reads OpenRouter remaining credits for openrouter, z.ai GLM Coding Plan token quota for zai (Global) and zai-coding-cn (BigModel China), and OpenAI Codex's ChatGPT subscription quota when using openai-codex OAuth; it skips platform API-key auth because that has no ChatGPT subscription quota. OpenRouter uses GET /api/v1/credits; z.ai uses GET /api/monitor/usage/quota/limit; Codex uses GET /wham/usage. z.ai and Codex show both each quota's window length and its nextResetTime/reset_at countdown (when supplied), and CODEX_API_URL or CHATGPT_BASE_URL can override the Codex base URL.

All providers share one renderer: every balance is projected to a list of segments and formatted the same way. Credits render as $1.5k; each usage window renders as [label ]<window> <remaining>%[ ↻<countdown>], e.g. 7d 72% ↻4d4h (↻ = resets in). Multiple windows are joined with ·, and named extra limits like Codex Spark get a label: 7d 72% ↻4d4h · Spark 7d 74% ↻5d4h. The footer refreshes on session start, model switch, and after each completed run (agent_settled), so it tracks both consumption and external tier changes for whatever provider is active — providers without a balance adapter are skipped, so this costs nothing for unrelated sessions. Do not also load the standalone kilo.ts or provider-balance.ts entries once this bundle is installed. If you previously symlinked the standalone provider-balance.ts, remove that link — the feature now ships in this bundle.