@diegopetrucci/pi-extensions 0.1.61 → 0.1.63

package/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/README.md

@@ -13,8 +13,9 @@ A collection of [pi](https://pi.dev) agent extensions I made. (btw, [I am also b
 
 ### Providers
 
-- [`claude-fast`](./extensions/claude-fast): Adds `/claude-fast` to enable Anthropic Claude Fast mode for supported Claude Opus models by injecting `speed: "fast"`.
-- [`openai-fast`](./extensions/openai-fast): Adds `/fast` to enable OpenAI Codex Fast mode for ChatGPT-auth GPT-5.4, GPT-5.5, and GPT-5.6 Codex variants by injecting the priority service tier.
+- [`claude-fast`](./extensions/claude-fast): Standalone-only provider-specific alternative that adds `/claude-fast` for supported Anthropic Claude Opus models.
+- [`fast`](./extensions/fast): Adds one `/fast` toggle that enables the provider-specific Fast mode for the currently selected supported OpenAI Codex or Anthropic Claude model.
+- [`openai-fast`](./extensions/openai-fast): Standalone-only provider-specific alternative that adds `/fast` for ChatGPT-auth GPT-5.4, GPT-5.5, and GPT-5.6 Codex variants.
 
 ### UI
 

package/extensions/agent-workflow-audit/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/extensions/agent-workflow-audit/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@diegopetrucci/pi-agent-workflow-audit",
-  "version": "0.1.8",
+  "version": "0.1.9",
   "description": "A pi extension that runs an isolated repo workflow audit and returns only the final report to the main session.",
   "keywords": [
     "pi-package",

package/extensions/annotate-git-diff/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/extensions/annotate-git-diff/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@diegopetrucci/pi-annotate-git-diff",
-  "version": "0.1.8",
+  "version": "0.1.9",
   "description": "A standalone pi extension that adds /annotate-git-diff, a native Glimpse UI for annotating git diffs and sending feedback to the editor.",
   "keywords": [
     "pi-package",

package/extensions/annotate-last-message/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/extensions/annotate-last-message/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@diegopetrucci/pi-annotate-last-message",
-  "version": "0.1.6",
+  "version": "0.1.7",
   "description": "A standalone pi extension that adds /annotate-last-message, a native Glimpse UI for annotating the latest assistant reply.",
   "keywords": [
     "pi-package",

package/extensions/brrr/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/extensions/brrr/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@diegopetrucci/pi-brrr",
-  "version": "0.1.11",
+  "version": "0.1.12",
   "description": "A pi extension that sends brrr push notifications when pi is ready for input.",
   "keywords": [
     "pi-package",

package/extensions/claude-fast/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/extensions/claude-fast/README.md

@@ -2,6 +2,8 @@
 
 A pi extension that enables Anthropic Claude Fast mode for supported Claude Opus models.
 
+This package is standalone-only and is not auto-loaded by the `@diegopetrucci/pi-extensions` collection package. The collection uses the unified [`fast`](../fast) extension instead.
+
 When active, the extension injects this into eligible Anthropic request payloads:
 
 ```json

package/extensions/claude-fast/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@diegopetrucci/pi-claude-fast",
-  "version": "0.1.11",
+  "version": "0.1.13",
   "description": "A pi extension that enables Anthropic Claude Fast mode for supported Claude Opus models by injecting speed=fast.",
   "keywords": [
     "pi-package",

package/extensions/code-reviewer/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/extensions/code-reviewer/README.md

@@ -44,6 +44,7 @@ When called, it launches an isolated in-memory child agent with:
 
 - no inherited extensions, skills, prompt templates, themes, context files, or agents files;
 - an auto-selected review model that prefers contrarian-style opposite-provider and opposite-family candidates before same-provider fallbacks;
+- automatic model candidates constrained to the current Pi session's model scope when that scope is non-empty;
 - requested thinking taken from the active session when available, otherwise defaulting to `high` for reasoning models and `off` for non-reasoning models, then clamped to the selected model's supported level;
 - read-only tools only: `read`, `grep`, `find`, `ls`, and guarded `bash`;
 - a local-checkout path guard for file inspection;

package/extensions/code-reviewer/index.ts

@@ -224,8 +224,28 @@ type PiModel = {
 type ModelRegistryContext = {
 	model?: PiModel;
 	modelRegistry: { getAvailable(): PiModel[] | Promise<PiModel[]> };
+	scopedModels?: ReadonlyArray<{ model: PiModel; thinkingLevel?: ThinkingLevel }>;
 };
 
+function modelKey(model: PiModel): string {
+	return `${model.provider}/${model.id}`.toLowerCase();
+}
+
+function hasModelScope(ctx: ModelRegistryContext): boolean {
+	return Array.isArray(ctx.scopedModels) && ctx.scopedModels.length > 0;
+}
+
+function isModelInScope(ctx: ModelRegistryContext, model: PiModel): boolean {
+	if (!hasModelScope(ctx)) return true;
+	const key = modelKey(model);
+	return ctx.scopedModels!.some((entry) => modelKey(entry.model) === key);
+}
+
+async function getAutoSelectableModels(ctx: ModelRegistryContext): Promise<PiModel[]> {
+	const available = await ctx.modelRegistry.getAvailable();
+	return hasModelScope(ctx) ? available.filter((model) => isModelInScope(ctx, model)) : available;
+}
+
 type CreateAgentSessionOptions = NonNullable<Parameters<typeof createAgentSession>[0]>;
 type CreateAgentSessionModel = CreateAgentSessionOptions["model"];
 
@@ -354,11 +374,13 @@ function appendOrderedCandidates(
 async function selectCodeReviewerModel(
 	ctx: ModelRegistryContext,
 ): Promise<{ ok: true; selection: PiModel; ordered: PiModel[] } | { ok: false; error: string }> {
-	const available = await ctx.modelRegistry.getAvailable();
+	const available = await getAutoSelectableModels(ctx);
 	if (available.length === 0) {
 		return {
 			ok: false,
-			error: "No authenticated models are available. Log in or configure an API key first.",
+			error: hasModelScope(ctx)
+				? "No authenticated models are available in the current session model scope. Adjust the scope, log in, or configure an API key."
+				: "No authenticated models are available. Log in or configure an API key first.",
 		};
 	}
 

package/extensions/code-reviewer/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@diegopetrucci/pi-code-reviewer",
-  "version": "0.1.6",
+  "version": "0.1.7",
   "description": "A standalone pi extension that adds a read-only code_reviewer subagent tool for isolated code reviews.",
   "keywords": [
     "pi-package",

package/extensions/confirm-destructive/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/extensions/confirm-destructive/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@diegopetrucci/pi-confirm-destructive",
-  "version": "0.1.9",
+  "version": "0.1.10",
   "description": "A pi extension that confirms destructive session actions.",
   "keywords": [
     "pi-package",

package/extensions/context-cap/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/extensions/context-cap/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@diegopetrucci/pi-context-cap",
-  "version": "0.1.8",
+  "version": "0.1.9",
   "description": "A pi extension that caps effective model context windows at 200k tokens for earlier auto-compaction.",
   "keywords": [
     "pi-package",

package/extensions/context-inspector/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/extensions/context-inspector/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@diegopetrucci/pi-context-inspector",
-  "version": "0.1.10",
+  "version": "0.1.11",
   "description": "A pi extension that opens a local HTML dashboard explaining where the current session context is going.",
   "keywords": [
     "pi-package",

package/extensions/contrarian/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/extensions/contrarian/README.md

@@ -77,6 +77,8 @@ Tool-call parameters still win over these defaults. `auto` clears the configured
 - `thinkingLevel` - optional reasoning/thinking override; falls back to the `/contrarian thinking` default, then built-in defaults
 - `cwd` - optional working directory override
 
+When the current Pi session has a non-empty model scope, Contrarian limits automatic model selection and automatic fallbacks to that scope. An explicit tool or `/contrarian model` override remains authoritative even when it is outside the session scope.
+
 ## Notes
 
 - The contrarian is intentionally **read-only**.

package/extensions/contrarian/index.ts

@@ -20,6 +20,31 @@ type PiModel = {
 	thinkingLevelMap?: ThinkingLevelMap;
 };
 
+type ModelSelectionContext = {
+	model?: PiModel;
+	modelRegistry: { getAvailable(): PiModel[] | Promise<PiModel[]> };
+	scopedModels?: ReadonlyArray<{ model: PiModel; thinkingLevel?: ThinkingLevel }>;
+};
+
+function modelKey(model: PiModel): string {
+	return `${model.provider}/${model.id}`.toLowerCase();
+}
+
+function hasModelScope(ctx: ModelSelectionContext): boolean {
+	return Array.isArray(ctx.scopedModels) && ctx.scopedModels.length > 0;
+}
+
+function isModelInScope(ctx: ModelSelectionContext, model: PiModel): boolean {
+	if (!hasModelScope(ctx)) return true;
+	const key = modelKey(model);
+	return ctx.scopedModels!.some((entry) => modelKey(entry.model) === key);
+}
+
+async function getAutoSelectableModels(ctx: ModelSelectionContext): Promise<PiModel[]> {
+	const available = await ctx.modelRegistry.getAvailable();
+	return hasModelScope(ctx) ? available.filter((model) => isModelInScope(ctx, model)) : available;
+}
+
 interface UsageStats {
 	input: number;
 	output: number;
@@ -180,8 +205,6 @@ const PROVIDER_MODEL_PREFERENCES: Record<string, string[]> = {
 		"accounts/fireworks/models/glm-5p2",
 		"accounts/fireworks/routers/glm-5p2-fast",
 		"accounts/fireworks/models/minimax-m3",
-		"accounts/fireworks/models/glm-5p1",
-		"accounts/fireworks/routers/glm-5p1-fast",
 		"accounts/fireworks/models/kimi-k2p6",
 		"accounts/fireworks/routers/kimi-k2p6-turbo",
 		"accounts/fireworks/routers/kimi-k2p6-fast",
@@ -918,7 +941,7 @@ function appendThinkingLevelClampReason(
 }
 
 async function findAvailableModel(
-	ctx: { model?: PiModel; modelRegistry: { getAvailable(): PiModel[] | Promise<PiModel[]> } },
+	ctx: ModelSelectionContext,
 	modelRef: string,
 ): Promise<PiModel | undefined> {
 	const available = await ctx.modelRegistry.getAvailable();
@@ -964,11 +987,11 @@ function toContrarianSelection(
 }
 
 function buildSessionFallbackSelection(
-	ctx: { model?: PiModel },
+	ctx: ModelSelectionContext,
 	thinkingLevelOverride: ThinkingLevel | undefined,
 ): ContrarianSelection | undefined {
 	const model = ctx.model;
-	if (!model) return undefined;
+	if (!model || !isModelInScope(ctx, model)) return undefined;
 	return toContrarianSelection(
 		model,
 		thinkingLevelOverride,
@@ -984,14 +1007,16 @@ function withFallbackReason(selection: ContrarianSelection, previous: Contrarian
 }
 
 async function selectContrarianModel(
-	ctx: { model?: PiModel; modelRegistry: { getAvailable(): PiModel[] | Promise<PiModel[]> } },
+	ctx: ModelSelectionContext,
 	thinkingLevelOverride?: ThinkingLevel,
 ): Promise<{ ok: true; selection: ContrarianSelection; ordered: ContrarianSelection[] } | { ok: false; error: string }> {
-	const available = await ctx.modelRegistry.getAvailable();
+	const available = await getAutoSelectableModels(ctx);
 	if (available.length === 0) {
 		return {
 			ok: false,
-			error: "No authenticated models are available. Log in or configure an API key first.",
+			error: hasModelScope(ctx)
+				? "No authenticated models are available in the current session model scope. Adjust the scope, log in, or configure an API key."
+				: "No authenticated models are available. Log in or configure an API key first.",
 		};
 	}
 
@@ -1340,6 +1365,7 @@ type ContrarianExtensionDeps = {
 };
 
 export const __test__ = {
+	buildSessionFallbackSelection,
 	createContrarianExtension,
 	findAvailableModel,
 	isModelAvailabilityError,

package/extensions/contrarian/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@diegopetrucci/pi-contrarian",
-  "version": "0.1.8",
+  "version": "0.1.9",
   "description": "A read-only contrarian subagent extension for pi that stress-tests plans, designs, assumptions, and conclusions.",
   "keywords": [
     "pi-package",

package/extensions/dirty-repo-guard/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/extensions/dirty-repo-guard/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@diegopetrucci/pi-dirty-repo-guard",
-  "version": "0.1.8",
+  "version": "0.1.9",
   "description": "A pi extension that prompts before session changes when the current git repo has uncommitted changes.",
   "keywords": [
     "pi-package",

package/extensions/dynamic-context-pruning/.pi-fleet-tested-version

@@ -1 +1 @@
-0.82.1
+0.83.0

package/extensions/dynamic-context-pruning/README.md

@@ -230,19 +230,32 @@ never required to exist. Full shape, with defaults:
 
 Notes:
 
-- `thresholds.minCharsSaved` is reserved for a future minimum-size filter and
-  is **not** currently enforced by the pipeline; the net-benefit gate's
-  break-even math is what actually decides whether a prune applies.
+- `thresholds.minCharsSaved` is a minimum-size floor enforced against fresh
+  **automatic** strategy proposals only, before the net-benefit gate runs: a
+  text-only candidate whose raw character savings (original text chars minus
+  placeholder chars) fall below this value is dropped up front and never
+  reaches gate or batch consideration. A tool-result candidate containing one
+  or more image blocks bypasses this raw-text floor (images do not receive
+  synthetic character counts) and is evaluated by the existing token-based
+  net-benefit gate. Manual prunes (`/prune`) and already-persisted/replayed
+  decisions always bypass this floor, same as the gate. Set to `0` to disable
+  filtering entirely. This is independent of the net-benefit gate's
+  break-even math (`gate.*`), which is a separate, later check on whatever
+  survives this floor.
 - `gate.breakEvenThresholdByState` lets `idle` vs `mid_loop` agent states use
   different thresholds; both default to `gate.breakEvenThreshold`. Real
-  mid-loop/idle detection isn't wired through yet, so this always evaluates as
-  `idle` in v1 — the field exists so the seam is ready once that lands. The
-  representative-corpus benchmark (see [Roadmap](#roadmap)) shows `mid_loop`
-  candidates carry ~all realized net benefit and `idle` candidates carry ~none
-  at r=0.1, which argues for a stricter `idle` default — but since every live
-  evaluation runs as `idle` today, defaulting it stricter would silently
-  disable most automatic pruning. Both states are intentionally kept at
-  parity until real agent-state detection lands; that's a follow-up ticket.
+  mid-loop/idle detection is wired through the `context` event handler
+  (pe-zy4s): the state is classified straight from the message payload —
+  `idle` if the most recent relevant message is a user message (this is the
+  first LLM call of the turn), `mid_loop` if it's an assistant message or a
+  tool result (the agent is iterating mid-turn). The representative-corpus
+  benchmark (see [Roadmap](#roadmap)) was re-derived against this same
+  runtime-observable definition and found the state-split REVERSED from an
+  earlier turn-end-based analysis: at r=0.1, `idle` candidates now carry
+  essentially all the realized net benefit and `mid_loop` candidates carry
+  essentially none. Since `idle` is not the worthless state under this
+  definition, both states are kept at parity (`22`/`22`) rather than forcing
+  a stricter default in either direction.
 - `gate.breakEvenThreshold`'s default of `22` is calibrated from the
   representative-corpus benchmark at `cachedPriceRatio` r=0.1 (aggressive
   prompt caching, the common case) — see [Roadmap](#roadmap) for the full
@@ -296,8 +309,46 @@ Options:
 - `--ratio R`: cached-price ratio(s) to model; repeatable and/or
   comma-separated (e.g. `--ratio 0.1,0.2`). Default: `0.1`.
 - `--json`: emit a full machine-readable JSON dump instead of aligned text.
+- `--simulate-compression`: additionally SIMULATE v2-style range compression
+  (docs/v2-design.md §1/§4 go/no-go evidence) alongside — never mixed into —
+  the deterministic results. See below.
+- `--sim-summary-fraction F`: summary size as a fraction of range tokens;
+  repeatable/comma-separated (e.g. `--sim-summary-fraction 0.1,0.15,0.3`).
+  Default: `0.15`.
+- `--sim-summary-min-tokens N`: floor on summary tokens regardless of
+  fraction. Default: `200`.
+- `--sim-summarizer-cost-mult M`: relative per-token price of the summarizer
+  call vs. the main model. Default: `1.0` (same per-token price; a
+  deliberate simplification that ignores summarizer latency/availability).
+- `--sim-min-range-tokens N`: minimum contiguous range size to be
+  considered. Default: `2000`.
 - `--help`: print usage.
 
+### Compression-simulation mode (`--simulate-compression`)
+
+This is a v2 go/no-go evidence generator that does **not** require building
+v2: it identifies plausible v2-style `compress` (range mode) opportunities —
+contiguous spans of complete, non-recent, non-protected messages, never
+splitting a toolCall/toolResult pair — during the same replay pass, and
+models the outcome:
+
+```text
+summaryTokens      = max(fraction * rangeTokens, summaryMinTokens)
+oneTimeCost        = cacheBustPenalty(earliest range position)
+                       + summarizerCostMult * (rangeTokens + summaryTokens)
+recurringSaving     = r * (rangeTokens - summaryTokens)
+realizedNetBenefit  = actualRemainingCalls * recurringSaving - oneTimeCost
+```
+
+Each distinct range is recorded once, at the earliest call boundary where it
+first exists and meets `--sim-min-range-tokens`; ranges never overlap. The
+same auto-expanding break-even threshold sweep as the deterministic
+strategies runs against this separate SIMULATED population, reported
+side-by-side with (but never combined with) the deterministic results — the
+two populations are computed **independently**; a simulated range may fully
+contain deterministic candidates, and no overlap/interaction between the two
+is modeled or netted out.
+
 With no positional paths, it defaults to every `*.jsonl` file found
 recursively under `~/.pi/agent/sessions`.
 
@@ -387,6 +438,22 @@ realized benefit across the whole 1,390-session corpus is only ~20.6k
 token-units — economically marginal, on the order of pennies. Savings only
 become material at weaker caching, r>=0.25.
 
+**pe-zy4s update (2026-07-08): this idle/mid_loop split used a turn-END
+definition** ("idle" = this call IS the turn's final assistant message, only
+knowable in hindsight via replay) **that was never runtime-observable.** Once
+real runtime agent-state detection landed (turn-START definition: "idle" =
+this is the FIRST LLM call of a turn; see the config-notes bullet above), the
+benchmark's candidate labeling was aligned to that same definition and
+re-derived on the representative corpus (460 gate-eligible candidates on the
+current `~/.the-last-harness/agent/sessions` corpus). The split **reverses**:
+at r=0.1, `idle` (T=22, ~20.6k realized net benefit) now carries essentially
+all the value, and `mid_loop` (T=1, ~0 realized net benefit) carries
+essentially none. Per the decision rule (stricter idle default only if idle
+is shown ~worthless), this evidence does **not** support a stricter idle
+default — parity (`22`/`22`) is kept. The table above is retained as the
+historical turn-END-definition record; it no longer reflects how the runtime
+classifies agent state.
+
 **This reframes, rather than weakens, the case for v2.** Small deterministic
 removals (dedupe/error-purge/superseded-file-ops) structurally cannot beat
 the cache-bust penalty at r=0.1 — there just isn't enough single-message