@czottmann/pi-automode 1.8.2 → 1.10.0

package/README.md

@@ -88,12 +88,25 @@ This is project-local and should not be committed. Shared project `.pi/automode.
 
 Set a global default classifier model in `~/.pi/agent/automode.json`; override it per project in `.pi/automode.local.json`.
 
+`classifierReasoningLevel` optionally requests `low`, `medium`, `high`, `xhigh`, or `max` reasoning for both classifier stages. If the key is absent, pi-automode sends no reasoning preference and leaves the choice to the server. Pi AI clamps unsupported values to the nearest level supported by the selected model; a non-reasoning model resolves to `off`. `low` matches Codex Auto Review's reasoning effort and the practical default when an explicit value is needed. Higher levels can consume the existing 512/1200-token stage limits before producing visible output, which causes the classifier to fail closed. Raise `fastClassifierMaxTokens` (default 512, integer ≥ 16) if you run a reasoning model whose fast-stage budget is truncated before it emits the required `0`/`1` digit.
+
+`allowInsideWorkingDirectory` (default `false`) adds a deterministic silent-allow tier for the file tools (`read`, `write`, `edit`, `grep`, `find`, `ls`): when `true`, a call whose resolved path is inside the working directory is allowed without any classifier call, and file access outside the working directory is routed to the classifier (including reads, which would otherwise take the read-only fast path). This matches the Codex/Claude Code "inside the sandbox = silent, outside = review" model. The tier takes precedence over `classifyReadOnlyTools`: with both enabled, in-tree file access is still allowed without a classifier call, and out-of-tree file access is classified. `classifyReadOnlyTools: true` only routes in-tree reads to the classifier when `allowInsideWorkingDirectory` is `false`.
+
+`deniedPaths` (default `[]`) is a list of path glob patterns that are hard-denied before the classifier and before the inside-working-directory tier — the file-tool equivalent of a secret/system deny list. Patterns support `~`, `$HOME`, and `${HOME}` expansion and `*` (which matches any characters, including `/`, so `**/id_rsa` matches a private key at any depth). Matching checks both the path as typed and its symlink-resolved form, so a `~/.ssh/*` rule still matches when `~/.ssh` is a symlink. A matching path blocks the call unconditionally (no classifier, no override). The deny list applies to file tools only; `bash` path access is governed by the classifier. Both keys follow the normal scalar/array precedence.
+
+The setting follows the normal scalar precedence: global, then project-local, then `PI_AUTOMODE_SETTINGS_JSON`. Shared project `.pi/automode.json` cannot set it. Omitting the key at a higher-precedence scope does not clear a lower-precedence value.
+
 Example:
 
 ```json
 {
   "autoMode": {
     "classifierModel": "provider/model-id",
+    "classifierReasoningLevel": "low",
+    "classifyReadOnlyTools": false,
+    "fastClassifierMaxTokens": 512,
+    "allowInsideWorkingDirectory": false,
+    "deniedPaths": [],
     "maxUserTranscriptTokens": 4000,
     "maxToolTranscriptTokens": 4000,
     "environment": [
@@ -120,6 +133,10 @@ Example:
 
 `maxUserTranscriptTokens` and `maxToolTranscriptTokens` are approximate per-category budgets; both default to 4000 and accept integers of at least 32. The former `maxTranscriptLines` setting is no longer supported because evidence selection is token-budgeted rather than line-based.
 
+### Ask-user tools and explicit authorization
+
+Classifier evidence includes normal user messages and assistant tool-call inputs, but excludes assistant prose and all tool results. This includes answers returned by ask-user tools such as `@vanillagreen/pi-questions`. Selecting "Yes" there helps the agent decide what to do next, but pi-automode does not treat that tool result as explicit authorization to override a soft deny. Send the authorization as a normal chat message instead; the agent can then retry the action. Tool results are excluded because they may contain untrusted or prompt-injected content.
+
 ### `$defaults`
 
 See [Defaults and rule-list behavior](docs/defaults.md) for built-in `environment`, `allow`, `protectedPaths`, `soft_deny`, and `hard_deny` entries, plus replacement behavior when `$defaults` is omitted.
@@ -160,7 +177,9 @@ The extension blocks these before any allow or classifier decision:
 - root, home, and system-path destructive deletes
 - edits to `.pi/automode*`, `.pi` auto-mode files, and this extension's safety-control files
 
-Read-only Pi tools (`read`, `grep`, `find`, `ls`) are allowed after those checks. Every side-effecting action goes to the classifier, including all `write` and `edit` calls, `bash`, MCP, subagent, network-capable tools, and unknown tools. This keeps classifier hard-deny rules unconditional; direct file writes cannot bypass them.
+Read-only Pi tools (`read`, `grep`, `find`, `ls`) are allowed after those checks. Every side-effecting action goes to the classifier, including all `write` and `edit` calls, `bash`, MCP, subagent, network-capable tools, and unknown tools. This keeps classifier hard-deny rules unconditional; direct file writes cannot bypass them. Set `classifyReadOnlyTools: true` to route read-only tools through the classifier as well, so reads outside the trusted working tree can be denied by policy. With it enabled, every `read`, `grep`, `find`, and `ls` call runs the two-stage classifier, which raises the number of model calls, the latency, and the cost per session.
+
+Path matches in `deniedPaths` are blocked before every classifier and fast-path decision, so secret and system paths never reach the model through the file tools. The deny list does not govern `bash`; shell access to those paths is handled by the classifier and the deterministic hard-deny checks. With `allowInsideWorkingDirectory: true`, file tools inside the working directory are allowed without a classifier call, and outside-working-directory file access (reads included) goes to the classifier.
 
 Classification starts with a one-token conservative filter and runs structured review only when that filter requests it. Both stages use a classifier-specific session key and short provider cache retention where the provider supports it. Missing models, provider failures, or malformed responses block the action.
 

package/docs/GLOSSARY.md

@@ -0,0 +1,35 @@
+# Glossary
+
+Canonical vocabulary for `@czottmann/pi-automode`. Project-specific terms only — standard technical words appear here only when this project uses them in a specific way. Longer explanations live in the docs linked from each entry.
+
+## Enforcement flow
+
+The ordered pipeline that runs on every agent tool call before execution. See [Auto-mode classifier flow](automode-classifier-flow.md).
+
+**Auto mode** — Claude Code-style guardrail posture: a pre-execution classifier allows routine, reversible actions and blocks risky ones, replacing routine permission prompts.
+
+**Deterministic hard-deny** — Local code checks that block high-risk actions before any classifier call and cannot be overridden. Distinct from the config-level [hard_deny](#classifier-policy-and-rules); independent of the model.
+
+**Read-only bypass** — The default fast path where the read-only tools (`read`, `grep`, `find`, `ls`) are allowed without classifier review once permission and deterministic checks pass. `classifyReadOnlyTools` routes them through the classifier instead.
+
+**Staged classifier** — The two-stage safety classifier: a conservative one-token filter gates an optional structured review. See [Fast stage](#enforcement-flow) and [Detailed stage](#enforcement-flow).
+
+**Fast stage** — The first classifier stage: a one-token filter that returns `0` (clearly allowed) or `1` (may need review).
+
+**Detailed stage** — The second classifier stage, run when the fast stage requests review, that returns a structured allow/block decision.
+
+## Classifier policy and rules
+
+The classifier's deny tiers and rule-list syntax. See [Defaults and rule-list behavior](defaults.md).
+
+**hard_deny** — Unconditional classifier rules that cannot be overridden. Distinct from the code-level [deterministic hard-deny](#enforcement-flow) checks.
+
+**soft_deny** — Overridable classifier block rules, unlike [hard_deny](#classifier-policy-and-rules).
+
+**explicit_intent** — A classifier tier meaning the allow was justified because the user's latest instruction directly and specifically authorized an otherwise [soft-denied](#classifier-policy-and-rules) action.
+
+**`$defaults`** — A section-local marker in a rule list that expands to the built-in entries for that list.
+
+## Status
+
+**AM status line** — The persistent TUI footer (prefixed `AM`) that reports auto-mode status: enabled/disabled and action/classifier counts.

package/docs/automode-classifier-flow.md

@@ -253,6 +253,10 @@ The classifier model is selected in this order:
 
 `/automode model provider/model-id` and the interactive model picker save `autoMode.classifierModel` to `~/.pi/agent/automode.json`. Project-local `.pi/automode.local.json` can still override that global choice.
 
+`autoMode.classifierReasoningLevel` can request `low`, `medium`, `high`, `xhigh`, or `max` reasoning for both classifier stages. When the key is absent, classifier calls use the raw completion path and omit a reasoning preference so the server can choose its default. When it is set, classifier calls use Pi AI's normalized completion path. Pi AI clamps the request to the nearest level supported by the model; non-reasoning models resolve to `off`, remain on the normalized path, and receive no reasoning preference.
+
+Reasoning does not raise the stage token limits. A high level can consume the fast stage's 512 tokens or the detailed stage's 1200 tokens before producing valid visible output. Truncation still fails closed. `low` is the practical explicit setting and matches Codex Auto Review.
+
 The extension asks Pi's model registry for API credentials. If the model cannot be found or credentials are unavailable, classification returns a blocking decision:
 
 ```text

package/docs/defaults.md

@@ -54,6 +54,10 @@ Protected files: `.gitconfig`, `.gitmodules`, `.gitignore`, `.gitattributes`, sh
 
 Read-only tools (`read`, `grep`, `find`, `ls`) remain locally allowed after permission and deterministic checks. Writes and edits always require classification, regardless of their target.
 
+### `deniedPaths`
+
+`deniedPaths` is a separate opt-in list (default `[]`, no built-in entries, so `$defaults` is a no-op; it is accepted for consistency with the other rule lists) of path glob patterns that are hard-denied for the file tools (`read`, `write`, `edit`, `grep`, `find`, `ls`) before any classifier or fast-path decision. Use it for secrets and system paths that must never reach the model through the file tools; `bash` access to such paths remains classifier-governed. Patterns support `~`/`$HOME`/`${HOME}` expansion and `*` (matches any characters, including `/`). Matching checks both the typed path and its symlink-resolved form. `deniedPaths` only restricts; it never grants access.
+
 ### `soft_deny`
 
 `$defaults` expands to soft blocks for:

package/docs/observability-logging.md

@@ -55,6 +55,21 @@ One per tool-call decision. Every allow and every block goes through exactly one
 | `outcome` | `allow` or `block` |
 | `reason` | the reason string (classifier reason, or the deterministic/permission reason) |
 | `classifierModel` | the configured classifier model, when relevant |
+| `reasoning` | classifier reasoning mode and requested/effective level; see below |
+
+The reasoning field records either server-default mode:
+
+```json
+{"mode":"server-default"}
+```
+
+or an explicit request after model-level clamping:
+
+```json
+{"mode":"explicit","requestedLevel":"max","effectiveLevel":"xhigh"}
+```
+
+Classifier-routed decisions contain the effective level once the configured model resolves, even when `classifierIo` is off or authentication then fails. If the configured model itself cannot be resolved, an explicit entry records `requestedLevel` without `effectiveLevel` because no model-supported level exists. A local permission, deterministic, or read-only decision does not run the classifier and likewise may omit `effectiveLevel`. In `server-default` mode, the concrete server-selected level is not observable and is not inferred.
 
 ### `message` (classifier usage)
 
@@ -78,6 +93,7 @@ Written only for classifier-routed actions, and only when `classifierIo: true`.
 | `ts` | ISO timestamp |
 | `decisionId` | matches the `decision` entry for the same call |
 | `model` | classifier model used, e.g. `anthropic/claude-haiku-4` |
+| `reasoning` | `server-default`, or the explicit requested and effective model-supported level |
 | `prompt.system` | the full system policy with `environment`/`allow`/`soft_deny`/`hard_deny` rules interpolated |
 | `prompt.context` | the shared context message: loaded project instructions + classifier transcript + action |
 | `prompt.fastInstruction` | the exact one-token filter instruction |

package/examples/automode.local.json

@@ -1,5 +1,12 @@
 {
   "autoMode": {
+    "classifierReasoningLevel": "low",
+    "allowInsideWorkingDirectory": false,
+    "deniedPaths": [
+      "*.env",
+      "~/.ssh/*",
+      "/etc/*"
+    ],
     "environment": [
       "$defaults",
       "Source control: github.example.com/acme-corp and all repos under it",

package/extensions/auto-mode/classifier.ts

@@ -1,8 +1,13 @@
 import { createHash } from "node:crypto";
-import { complete } from "@earendil-works/pi-ai";
+import { clampThinkingLevel } from "@earendil-works/pi-ai";
+import {
+  complete,
+  completeSimple,
+} from "@earendil-works/pi-ai/compat";
 import type {
   AssistantMessage,
   Model,
+  ProviderHeaders,
   UserMessage,
 } from "@earendil-works/pi-ai";
 import type { ExtensionContext } from "@earendil-works/pi-coding-agent";
@@ -10,6 +15,7 @@ import {
   CLASSIFIER_DETAILED_INSTRUCTION,
   CLASSIFIER_FAST_INSTRUCTION,
   CLASSIFIER_SYSTEM_PROMPT,
+  DEFAULT_FAST_CLASSIFIER_MAX_TOKENS,
 } from "./constants.ts";
 import { formatModelSpec, parseModelSpec } from "./model.ts";
 import { buildClassifierTranscript } from "./transcript.ts";
@@ -17,7 +23,11 @@ import type {
   ClassificationDecision,
   ClassifyAction,
   ClassifierIoAttempt,
+  ClassifierReasoning,
+  ClassifierReasoningLevel,
+  ClassifierReasoningLog,
   ClassifyResult,
+  EffectiveClassifierReasoningLevel,
   EffectiveConfig,
 } from "./types.ts";
 
@@ -40,13 +50,28 @@ export function buildClassifierPrompt(config: EffectiveConfig): string {
     );
 }
 
+type ClassifierResolution = {
+  reasoning: ClassifierReasoningLog;
+  classifier?: {
+    model: Model<any>;
+    apiKey?: string;
+    headers?: ProviderHeaders;
+  };
+  completionPlan?: ClassifierCompletionPlan;
+};
+
+export function classifierReasoningForConfig(
+  requestedLevel: ClassifierReasoningLevel | undefined,
+): ClassifierReasoningLog {
+  return requestedLevel === undefined
+    ? { mode: "server-default" }
+    : { mode: "explicit", requestedLevel };
+}
+
 async function resolveClassifier(
   ctx: ExtensionContext,
   config: EffectiveConfig,
-): Promise<
-  | { model: Model<any>; apiKey?: string; headers?: Record<string, string> }
-  | undefined
-> {
+): Promise<ClassifierResolution> {
   const configured = config.classifierModel;
   const model = configured
     ? (() => {
@@ -56,10 +81,27 @@ async function resolveClassifier(
         : undefined;
     })()
     : ctx.model;
-  if (!model) return undefined;
+  if (!model) {
+    return {
+      reasoning: classifierReasoningForConfig(config.classifierReasoningLevel),
+    };
+  }
+
+  const completionPlan = createClassifierCompletionPlan(
+    model,
+    config.classifierReasoningLevel,
+  );
   const auth = await ctx.modelRegistry.getApiKeyAndHeaders(model);
-  if (!auth.ok) return undefined;
-  return { model, apiKey: auth.apiKey, headers: auth.headers };
+  if (!auth.ok) return { reasoning: completionPlan.reasoning };
+  return {
+    reasoning: completionPlan.reasoning,
+    classifier: {
+      model,
+      apiKey: auth.apiKey,
+      headers: auth.headers,
+    },
+    completionPlan,
+  };
 }
 
 export type ClassifierCompletionFn = (
@@ -67,10 +109,11 @@ export type ClassifierCompletionFn = (
   options: { systemPrompt: string; messages: UserMessage[] },
   callOptions: {
     apiKey?: string;
-    headers?: Record<string, string>;
+    headers?: ProviderHeaders;
     signal?: AbortSignal;
     maxTokens: number;
     temperature?: number;
+    reasoning?: Exclude<EffectiveClassifierReasoningLevel, "off">;
     sessionId?: string;
     cacheRetention?: "none" | "short" | "long";
   },
@@ -80,6 +123,7 @@ export type RetryOptions = {
   maxAttempts?: number;
   maxTokens?: number;
   temperature?: number;
+  reasoningLevel?: Exclude<EffectiveClassifierReasoningLevel, "off">;
   sessionId?: string;
   cacheRetention?: "none" | "short" | "long";
   stage?: "fast" | "detailed";
@@ -87,13 +131,50 @@ export type RetryOptions = {
   onAttempt?: (attempt: ClassifierIoAttempt) => void;
 };
 
-const FAST_CLASSIFIER_MAX_TOKENS = 512;
-
 export type StagedClassifierOptions = {
   sessionId: string;
+  /** Override the fast-stage token budget; falls back to the default (512). */
+  fastClassifierMaxTokens?: number;
+  reasoningLevel?: Exclude<EffectiveClassifierReasoningLevel, "off">;
   onAttempt?: (attempt: ClassifierIoAttempt) => void;
 };
 
+export type ClassifierCompletionPlan = {
+  completeFn: ClassifierCompletionFn;
+  reasoning: ClassifierReasoning;
+  reasoningLevel?: Exclude<EffectiveClassifierReasoningLevel, "off">;
+};
+
+/** Select the raw or normalized Pi AI completion path and record the effective level. */
+export function createClassifierCompletionPlan(
+  model: Model<any>,
+  requestedLevel: ClassifierReasoningLevel | undefined,
+  rawComplete: ClassifierCompletionFn = complete,
+  simpleComplete: ClassifierCompletionFn = completeSimple,
+): ClassifierCompletionPlan {
+  if (requestedLevel === undefined) {
+    return {
+      completeFn: rawComplete,
+      reasoning: { mode: "server-default" },
+    };
+  }
+
+  const effectiveLevel = clampThinkingLevel(model, requestedLevel);
+  const reasoning: ClassifierReasoning = {
+    mode: "explicit",
+    requestedLevel,
+    effectiveLevel,
+  };
+  if (effectiveLevel === "off") {
+    return { completeFn: simpleComplete, reasoning };
+  }
+  return {
+    completeFn: simpleComplete,
+    reasoning,
+    reasoningLevel: effectiveLevel,
+  };
+}
+
 /** Concatenate all text blocks of an assistant message into a single string. */
 function extractAssistantText(message: AssistantMessage, trim = true): string {
   const text = message.content
@@ -224,7 +305,7 @@ export async function classifyWithRetry(
   classifier: {
     model: Model<any>;
     apiKey?: string;
-    headers?: Record<string, string>;
+    headers?: ProviderHeaders;
   },
   prompt: { systemPrompt: string; messages: UserMessage[] },
   signal: AbortSignal | undefined,
@@ -250,6 +331,9 @@ export async function classifyWithRetry(
           signal,
           maxTokens,
           ...(temperature === undefined ? {} : { temperature }),
+          ...(options.reasoningLevel === undefined
+            ? {}
+            : { reasoning: options.reasoningLevel }),
           sessionId: options.sessionId,
           cacheRetention: options.cacheRetention,
         },
@@ -292,7 +376,7 @@ export async function classifyInStages(
   classifier: {
     model: Model<any>;
     apiKey?: string;
-    headers?: Record<string, string>;
+    headers?: ProviderHeaders;
   },
   prompt: { systemPrompt: string; contextMessage: UserMessage },
   signal: AbortSignal | undefined,
@@ -316,7 +400,11 @@ export async function classifyInStages(
         signal,
         // Reasoning and OpenAI-compatible models may consume hidden reasoning,
         // control, and EOS tokens before emitting the required visible digit.
-        maxTokens: FAST_CLASSIFIER_MAX_TOKENS,
+        maxTokens: options.fastClassifierMaxTokens ??
+          DEFAULT_FAST_CLASSIFIER_MAX_TOKENS,
+        ...(options.reasoningLevel === undefined
+          ? {}
+          : { reasoning: options.reasoningLevel }),
         sessionId: options.sessionId,
         cacheRetention: "short",
       },
@@ -380,6 +468,7 @@ export async function classifyInStages(
       stage: "detailed",
       sessionId: options.sessionId,
       cacheRetention: "short",
+      reasoningLevel: options.reasoningLevel,
       onAttempt: options.onAttempt,
     },
   );
@@ -398,14 +487,17 @@ export const defaultClassifyAction: ClassifyAction = async (
   action,
   loadedContext,
 ): Promise<ClassifyResult> => {
-  const classifier = await resolveClassifier(ctx, config);
-  if (!classifier) {
+  const resolution = await resolveClassifier(ctx, config);
+  if (!resolution.classifier || !resolution.completionPlan) {
     return {
       decision: "block",
       tier: "none",
       reason: "No classifier model/API key available; auto mode fails closed.",
+      reasoning: resolution.reasoning,
     };
   }
+  const classifier = resolution.classifier;
+  const completionPlan = resolution.completionPlan;
 
   const systemPrompt = buildClassifierPrompt(config);
   const transcript = buildClassifierTranscript(ctx, {
@@ -426,20 +518,24 @@ export const defaultClassifyAction: ClassifyAction = async (
   const attempts: ClassifierIoAttempt[] = [];
   const started = Date.now();
   const decision = await classifyInStages(
-    complete,
+    completionPlan.completeFn,
     classifier,
     { systemPrompt, contextMessage },
     ctx.signal,
     {
       sessionId: classifierCacheSessionId(ctx),
+      fastClassifierMaxTokens: config.fastClassifierMaxTokens,
+      reasoningLevel: completionPlan.reasoningLevel,
       onAttempt: (attempt) => attempts.push(attempt),
     },
   );
 
   return {
     ...decision,
+    reasoning: completionPlan.reasoning,
     io: {
       model: formatModelSpec(classifier.model),
+      reasoning: completionPlan.reasoning,
       prompt: {
         system: systemPrompt,
         context: contextText,

package/extensions/auto-mode/config.ts

@@ -2,7 +2,11 @@ import { existsSync, mkdirSync, readFileSync, writeFileSync } from "node:fs";
 import { dirname, resolve } from "node:path";
 import {
   DEFAULT_ALLOW,
+  DEFAULT_ALLOW_INSIDE_WORKING_DIRECTORY,
+  DEFAULT_CLASSIFY_READ_ONLY_TOOLS,
+  DEFAULT_DENIED_PATHS,
   DEFAULT_ENVIRONMENT,
+  DEFAULT_FAST_CLASSIFIER_MAX_TOKENS,
   DEFAULT_HARD_DENY,
   DEFAULT_LOG_CONFIG,
   DEFAULT_MAX_TOOL_TRANSCRIPT_TOKENS,
@@ -16,6 +20,7 @@ import {
 import { parseToolPattern } from "./permissions.ts";
 import type {
   AutoModeSettings,
+  ClassifierReasoningLevel,
   ConfigLoadResult,
   EffectiveConfig,
   LoadedSettingsFile,
@@ -97,6 +102,11 @@ export function validateSettingsFile(
       const knownAutoMode = new Set([
         "enabled",
         "classifierModel",
+        "classifierReasoningLevel",
+        "classifyReadOnlyTools",
+        "fastClassifierMaxTokens",
+        "allowInsideWorkingDirectory",
+        "deniedPaths",
         "maxUserTranscriptTokens",
         "maxToolTranscriptTokens",
         "environment",
@@ -126,6 +136,44 @@ export function validateSettingsFile(
           `${source}: autoMode.classifierModel must be a provider/model string`,
         );
       }
+      if (
+        hasOwn(autoMode, "classifierReasoningLevel") &&
+        !isClassifierReasoningLevel(autoMode.classifierReasoningLevel)
+      ) {
+        diagnostics.push(
+          `${source}: autoMode.classifierReasoningLevel must be one of low, medium, high, xhigh, max`,
+        );
+      }
+      if (
+        hasOwn(autoMode, "classifyReadOnlyTools") &&
+        typeof autoMode.classifyReadOnlyTools !== "boolean"
+      ) {
+        diagnostics.push(
+          `${source}: autoMode.classifyReadOnlyTools must be a boolean`,
+        );
+      }
+      if (
+        hasOwn(autoMode, "fastClassifierMaxTokens") &&
+        (!Number.isInteger(autoMode.fastClassifierMaxTokens) ||
+          (autoMode.fastClassifierMaxTokens as number) < 16)
+      ) {
+        diagnostics.push(
+          `${source}: autoMode.fastClassifierMaxTokens must be an integer of at least 16`,
+        );
+      }
+      if (
+        hasOwn(autoMode, "allowInsideWorkingDirectory") &&
+        typeof autoMode.allowInsideWorkingDirectory !== "boolean"
+      ) {
+        diagnostics.push(
+          `${source}: autoMode.allowInsideWorkingDirectory must be a boolean`,
+        );
+      }
+      validateDeniedPathsSetting(
+        autoMode.deniedPaths,
+        source,
+        diagnostics,
+      );
       for (
         const key of [
           "maxUserTranscriptTokens",
@@ -273,10 +321,71 @@ function mergeLog(
   };
 }
 
+/**
+ * Validate `deniedPaths`: an array of non-empty path patterns. Unlike the
+ * `$defaults` rule lists there is no built-in default list, so `$defaults` is
+ * a no-op (accepted for consistency with the other rule lists) and omitting
+ * it is not a diagnostic.
+ */
+function validateDeniedPathsSetting(
+  value: unknown,
+  source: string,
+  diagnostics: string[],
+): void {
+  if (value === undefined) return;
+  if (!Array.isArray(value)) {
+    diagnostics.push(`${source}: deniedPaths must be an array of strings`);
+    return;
+  }
+  for (const [index, entry] of value.entries()) {
+    if (entry === "$defaults") continue;
+    if (typeof entry !== "string" || entry.trim() === "") {
+      diagnostics.push(
+        `${source}: deniedPaths[${index}] must be a non-empty path pattern`,
+      );
+      continue;
+    }
+    if (!DENIED_PATH_PATTERN_PREFIX.test(entry)) {
+      diagnostics.push(
+        `${source}: deniedPaths[${index}] "${entry}" can never match a resolved absolute path; start it with *, ~, $HOME, \${HOME}, or / (e.g. "**/${entry}")`,
+      );
+    }
+  }
+}
+
+/**
+ * A pattern can only match a resolved absolute path when it starts with a
+ * form that anchors it: a leading `/`, a home expansion (`~`, `$HOME`,
+ * `${HOME}`), or a `*` wildcard that absorbs the leading slash. Anything else
+ * (e.g. `config.json` or `src/secret.txt`) matches only against the bare
+ * relative name, which the matcher never sees.
+ */
+const DENIED_PATH_PATTERN_PREFIX =
+  /^(?:\/|~(?:\/|$)|\$HOME(?:\/|$)|\$\{HOME\}(?:\/|$)|\*)/;
+
+const CLASSIFIER_REASONING_LEVELS = new Set<ClassifierReasoningLevel>([
+  "low",
+  "medium",
+  "high",
+  "xhigh",
+  "max",
+]);
+
+export function isClassifierReasoningLevel(
+  value: unknown,
+): value is ClassifierReasoningLevel {
+  return typeof value === "string" &&
+    CLASSIFIER_REASONING_LEVELS.has(value as ClassifierReasoningLevel);
+}
+
 function validTranscriptBudget(value: unknown): value is number {
   return Number.isInteger(value) && Number(value) >= 32;
 }
 
+function validFastClassifierBudget(value: unknown): value is number {
+  return Number.isInteger(value) && Number(value) >= 16;
+}
+
 function applyAutoModeScalars(
   base: EffectiveConfig,
   settings: AutoModeSettings | undefined,
@@ -286,6 +395,20 @@ function applyAutoModeScalars(
     ...base,
     enabled: settings.enabled ?? base.enabled,
     classifierModel: settings.classifierModel ?? base.classifierModel,
+    classifierReasoningLevel: isClassifierReasoningLevel(
+        settings.classifierReasoningLevel,
+      )
+      ? settings.classifierReasoningLevel
+      : base.classifierReasoningLevel,
+    classifyReadOnlyTools: settings.classifyReadOnlyTools ??
+      base.classifyReadOnlyTools,
+    allowInsideWorkingDirectory:
+      settings.allowInsideWorkingDirectory ?? base.allowInsideWorkingDirectory,
+    fastClassifierMaxTokens: validFastClassifierBudget(
+        settings.fastClassifierMaxTokens,
+      )
+      ? settings.fastClassifierMaxTokens
+      : base.fastClassifierMaxTokens,
     maxUserTranscriptTokens: validTranscriptBudget(
         settings.maxUserTranscriptTokens,
       )
@@ -327,6 +450,10 @@ export function buildEffectiveConfigFromSources(
 ): EffectiveConfig {
   let config: EffectiveConfig = {
     enabled: true,
+    classifyReadOnlyTools: DEFAULT_CLASSIFY_READ_ONLY_TOOLS,
+    allowInsideWorkingDirectory: DEFAULT_ALLOW_INSIDE_WORKING_DIRECTORY,
+    deniedPaths: [...DEFAULT_DENIED_PATHS],
+    fastClassifierMaxTokens: DEFAULT_FAST_CLASSIFIER_MAX_TOKENS,
     maxUserTranscriptTokens: DEFAULT_MAX_USER_TRANSCRIPT_TOKENS,
     maxToolTranscriptTokens: DEFAULT_MAX_TOOL_TRANSCRIPT_TOKENS,
     environment: [...DEFAULT_ENVIRONMENT],
@@ -352,6 +479,7 @@ export function buildEffectiveConfigFromSources(
   const environment = createRuleAccumulator(DEFAULT_ENVIRONMENT);
   const allow = createRuleAccumulator(DEFAULT_ALLOW);
   const protectedPaths = createRuleAccumulator(DEFAULT_PROTECTED_PATHS);
+  const deniedPaths = createRuleAccumulator(DEFAULT_DENIED_PATHS);
   const softDeny = createRuleAccumulator(DEFAULT_SOFT_DENY);
   const hardDeny = createRuleAccumulator(DEFAULT_HARD_DENY);
 
@@ -360,6 +488,7 @@ export function buildEffectiveConfigFromSources(
     applyRuleSetting(environment, settings.autoMode?.environment);
     applyRuleSetting(allow, settings.autoMode?.allow);
     applyRuleSetting(protectedPaths, settings.autoMode?.protectedPaths);
+    applyRuleSetting(deniedPaths, settings.autoMode?.deniedPaths);
     applyRuleSetting(
       softDeny,
       settings.autoMode?.soft_deny ?? settings.autoMode?.softDeny,
@@ -375,6 +504,7 @@ export function buildEffectiveConfigFromSources(
     environment: finalizeRuleSetting(environment),
     allow: finalizeRuleSetting(allow),
     protectedPaths: finalizeRuleSetting(protectedPaths),
+    deniedPaths: finalizeRuleSetting(deniedPaths),
     softDeny: finalizeRuleSetting(softDeny),
     hardDeny: finalizeRuleSetting(hardDeny),
   };

package/extensions/auto-mode/constants.ts

@@ -155,6 +155,7 @@ export const CLASSIFIER_FAST_INSTRUCTION =
 export const CLASSIFIER_DETAILED_INSTRUCTION =
   `Return only JSON exactly matching:
 {"decision":"allow"|"block","tier":"hard_deny"|"soft_deny"|"allow"|"explicit_intent"|"none","reason":"brief concrete reason"}
+Do not use Markdown, code fences, prose, or any wrapper. The first character must be { and the last character must be }.
 Valid decision/tier combinations:
 - allow: allow, explicit_intent, or none
 - block: hard_deny, soft_deny, or none
@@ -178,6 +179,38 @@ export const PROFILE_FILES = new Set([
 
 export const READ_ONLY_TOOLS = new Set(["read", "grep", "find", "ls"]);
 
+/** File tools that expose a target path via `input.path` (for path gating). */
+export const PATH_BEARING_TOOLS = new Set([
+  "read",
+  "write",
+  "edit",
+  "grep",
+  "find",
+  "ls",
+]);
+
+/**
+ * Default behavior: read-only tools bypass the classifier entirely (the
+ * original auto-mode fast path). When set to true via config, read-only tools
+ * are routed through the classifier like any other action, so e.g. reads
+ * outside the trusted working tree can be denied by policy.
+ */
+export const DEFAULT_CLASSIFY_READ_ONLY_TOOLS = false;
+
+/** Default upper bound on fast-stage completion tokens (see PR note). */
+export const DEFAULT_FAST_CLASSIFIER_MAX_TOKENS = 512;
+
+/**
+ * Default behavior: no deterministic inside-working-directory tier; every
+ * non-read-only action is classified. When enabled, file tools whose resolved
+ * path is inside the working directory are allowed deterministically, and
+ * outside-CWD file access is routed to the classifier.
+ */
+export const DEFAULT_ALLOW_INSIDE_WORKING_DIRECTORY = false;
+
+/** Default path deny list: empty (no built-in secrets list). */
+export const DEFAULT_DENIED_PATHS: string[] = [];
+
 /** Default observability log config: off, classifier I/O off. */
 export const DEFAULT_LOG_CONFIG = {
   enabled: false,

package/extensions/auto-mode/extension.ts

@@ -3,7 +3,10 @@ import type {
   ExtensionCommandContext,
   ExtensionContext,
 } from "@earendil-works/pi-coding-agent";
-import { defaultClassifyAction } from "./classifier.ts";
+import {
+  classifierReasoningForConfig,
+  defaultClassifyAction,
+} from "./classifier.ts";
 import {
   AUTO_MODE_GUIDANCE,
   DEFAULT_ALLOW,
@@ -11,6 +14,7 @@ import {
   DEFAULT_HARD_DENY,
   DEFAULT_PROTECTED_PATHS,
   DEFAULT_SOFT_DENY,
+  PATH_BEARING_TOOLS,
   READ_ONLY_TOOLS,
 } from "./constants.ts";
 import {
@@ -27,7 +31,15 @@ import {
 } from "./log.ts";
 import { formatModelSpec, parseModelSpec } from "./model.ts";
 import { promptForClassifierModel } from "./model-selector.ts";
-import { matchesToolPattern } from "./permissions.ts";
+import { matchesDeniedPath, matchesToolPattern } from "./permissions.ts";
+import {
+  expandHomePattern,
+  extractInputPath,
+  isInside,
+  isProtectedPath,
+  resolveInputPath,
+  resolvePathForPolicy,
+} from "./paths.ts";
 import {
   actionSummary,
   formatDenials,
@@ -39,6 +51,7 @@ import {
 import { loadedContextFromSystemPromptOptions } from "./transcript.ts";
 import type {
   AutoModeState,
+  ClassifierReasoningLog,
   ClassifyAction,
   ClassifyResult,
   ConfigLoadResult,
@@ -61,10 +74,13 @@ type LogCtx = {
   logger: Logger;
   decisionId: string;
   classifierModel?: string;
+  reasoning: ClassifierReasoningLog;
 };
 
 /** Append ccusage-compatible usage and optional classifier I/O entries. */
 function logClassifierIo(decision: ClassifyResult, log: LogCtx): void {
+  if (decision.reasoning) log.reasoning = decision.reasoning;
+  if (decision.io) log.reasoning = decision.io.reasoning;
   if (!log.logger.enabled || !decision.io) return;
 
   for (const attempt of decision.io.attempts) {
@@ -87,6 +103,7 @@ function logClassifierIo(decision: ClassifyResult, log: LogCtx): void {
     ts: new Date().toISOString(),
     decisionId: log.decisionId,
     model: decision.io.model,
+    reasoning: decision.io.reasoning,
     prompt: decision.io.prompt,
     attempts: decision.io.attempts,
     durationMs: decision.io.durationMs,
@@ -170,6 +187,7 @@ export function createPiAutomode(options: PiAutomodeOptions = {}) {
           outcome: "block",
           reason: denial.reason,
           classifierModel: logCtx.classifierModel,
+          reasoning: logCtx.reasoning,
         });
       }
       if (ctx.hasUI) {
@@ -206,6 +224,7 @@ export function createPiAutomode(options: PiAutomodeOptions = {}) {
           outcome: "allow",
           reason,
           classifierModel: logCtx.classifierModel,
+          reasoning: logCtx.reasoning,
         });
       }
       return undefined;
@@ -232,7 +251,7 @@ export function createPiAutomode(options: PiAutomodeOptions = {}) {
       // Enforcement order:
       // 1. permission deny/ask rules,
       // 2. deterministic hard-deny checks that never consult the model,
-      // 3. read-only built-in fast path,
+      // 3. read-only built-in fast path (skipped when classifyReadOnlyTools is set),
       // 4. classifier for every remaining action, fail-closed on setup/parse errors.
       const cfg = effectiveConfig();
       if (!cfg.enabled) return undefined;
@@ -251,6 +270,7 @@ export function createPiAutomode(options: PiAutomodeOptions = {}) {
         }),
         decisionId: newDecisionId(),
         classifierModel: cfg.classifierModel,
+        reasoning: classifierReasoningForConfig(cfg.classifierReasoningLevel),
       };
 
       for (const pattern of cfg.permissionDeny) {
@@ -310,7 +330,73 @@ export function createPiAutomode(options: PiAutomodeOptions = {}) {
         }, logCtx);
       }
 
-      if (READ_ONLY_TOOLS.has(event.toolName)) {
+      // Deterministic path gate for file tools.
+      //
+      // `deniedPaths` always applies: a matching path is hard-denied before any
+      // classifier or fast path, so secrets and system dirs never reach the
+      // model. `allowInsideWorkingDirectory` adds a deterministic silent-allow
+      // tier for file tools whose resolved path is inside the working
+      // directory, and routes outside-CWD file access to the classifier
+      // (bypassing the read-only fast path so reads outside the tree are
+      // reviewed too).
+      //
+      // The gate is skipped entirely when both features are off, so the
+      // default configuration costs no extra filesystem calls.
+      let readOnlyFastPath =
+        !cfg.classifyReadOnlyTools && READ_ONLY_TOOLS.has(event.toolName);
+      if (
+        (cfg.deniedPaths.length > 0 || cfg.allowInsideWorkingDirectory) &&
+        PATH_BEARING_TOOLS.has(event.toolName)
+      ) {
+        const inputPath = extractInputPath(event.toolName, input);
+        if (inputPath !== undefined) {
+          const expanded = expandHomePattern(inputPath);
+          const resolved = resolveInputPath(ctx.cwd, expanded) ?? expanded;
+          const policyPath = resolvePathForPolicy(resolved) ?? resolved;
+          const denied =
+            cfg.deniedPaths.length > 0 &&
+            (matchesDeniedPath(resolved, cfg.deniedPaths) ||
+              matchesDeniedPath(policyPath, cfg.deniedPaths));
+          if (denied) {
+            return block(ctx, {
+              timestamp: Date.now(),
+              toolName: event.toolName,
+              reason: `Path denied by policy: ${policyPath}`,
+              action: summary,
+              kind: "deterministic-path-deny",
+            }, logCtx);
+          }
+          if (cfg.allowInsideWorkingDirectory) {
+            const policyCwd = resolvePathForPolicy(ctx.cwd) ?? ctx.cwd;
+            if (isInside(policyPath, policyCwd)) {
+              // Protected in-tree writes must still reach the classifier;
+              // otherwise the allow tier bypasses the protected-path policy
+              // for sensitive repository content such as .git/hooks/*, .pi/*,
+              // .husky/*, or .gitignore.
+              if (
+                (event.toolName === "write" || event.toolName === "edit") &&
+                isProtectedPath(policyPath, policyCwd, cfg.protectedPaths)
+              ) {
+                readOnlyFastPath = false;
+              } else {
+                return allow(
+                  ctx,
+                  "inside-working-directory",
+                  `Path inside working directory: ${policyPath}`,
+                  event.toolName,
+                  summary,
+                  logCtx,
+                );
+              }
+            }
+            // Outside the working directory: the read-only fast path must not
+            // apply; the classifier reviews this call.
+            readOnlyFastPath = false;
+          }
+        }
+      }
+
+      if (readOnlyFastPath) {
         return allow(
           ctx,
           "read-only",

package/extensions/auto-mode/hard-deny.ts

@@ -156,7 +156,18 @@ function isRecursiveRmArg(arg: string): boolean {
   );
 }
 
-function isRootHomeOrSystemPath(path: string): boolean {
+/**
+ * True for `/`, the user's home root, or a top-level system root such as
+ * `/etc`, `/usr`, or `/var`. Excludes the home *subtree*.
+ *
+ * On some distros (e.g. Fedora Silverblue) HOME lives under `/var`, which is
+ * in `systemRoots`. Without the subtree exemption, `path.startsWith("/var/")`
+ * would treat every path under HOME as a system root and hard-deny routine
+ * `rm -rf ~/...`. HOME itself is still matched below, so `rm -rf ~` stays
+ * blocked. `home` is a parameter so this can be unit-tested with a synthetic
+ * `/var/home/...` value.
+ */
+export function isRootHomeOrSystemPath(path: string, home: string): boolean {
   const systemRoots = [
     "/bin",
     "/boot",
@@ -170,9 +181,10 @@ function isRootHomeOrSystemPath(path: string): boolean {
     "/usr",
     "/var",
   ];
+  if (path.startsWith(`${home}/`)) return false;
   return (
     path === "/" ||
-    path === HOME ||
+    path === home ||
     systemRoots.some((root) => path === root || path.startsWith(`${root}/`))
   );
 }
@@ -261,7 +273,7 @@ function segmentHardDeny(
   if (name === "rm" && args.some(isRecursiveRmArg)) {
     for (const arg of args.filter((arg) => !arg.startsWith("-"))) {
       const path = shellPathTokenToPath(arg, cwd);
-      if (path && isRootHomeOrSystemPath(path)) {
+      if (path && isRootHomeOrSystemPath(path, HOME)) {
         return "irreversible deletion of home/root/system paths is hard-denied";
       }
     }
@@ -269,7 +281,7 @@ function segmentHardDeny(
 
   if (name === "find" && lowerArgs.includes("-delete")) {
     const root = shellPathTokenToPath(args[0] ?? "", cwd);
-    if (root && isRootHomeOrSystemPath(root) && root !== HOME) {
+    if (root && isRootHomeOrSystemPath(root, HOME) && root !== HOME) {
       return "system-wide delete is hard-denied";
     }
   }

package/extensions/auto-mode/log.ts

@@ -4,6 +4,7 @@ import { basename, dirname, extname, join } from "node:path";
 import type {
   ClassifierIo,
   ClassifierIoAttempt,
+  ClassifierReasoningLog,
   ClassificationDecision,
   DecisionKind,
 } from "./types.ts";
@@ -21,6 +22,7 @@ export type DecisionLogEntry = {
   outcome: "allow" | "block";
   reason: string;
   classifierModel?: string;
+  reasoning: ClassifierReasoningLog;
 };
 
 /** The classifier prompt, raw responses, and parsed decision for one action. */
@@ -29,6 +31,7 @@ export type ClassifierLogEntry = {
   ts: string;
   decisionId: string;
   model: string;
+  reasoning: ClassifierIo["reasoning"];
   prompt: ClassifierIo["prompt"];
   attempts: ClassifierIoAttempt[];
   durationMs: number;

package/extensions/auto-mode/paths.ts

@@ -7,7 +7,7 @@ import {
   relative,
   resolve,
 } from "node:path";
-import { HOME, PROFILE_FILES } from "./constants.ts";
+import { HOME, PATH_BEARING_TOOLS, PROFILE_FILES } from "./constants.ts";
 
 function stripLeadingAt(value: string): string {
   return value.startsWith("@") ? value.slice(1) : value;
@@ -22,6 +22,28 @@ export function resolveInputPath(
   return isAbsolute(raw) ? resolve(raw) : resolve(cwd, raw);
 }
 
+/** The target path a file tool operates on, from `input.path` (or undefined). */
+export function extractInputPath(
+  toolName: string,
+  input: Record<string, unknown>,
+): string | undefined {
+  if (!PATH_BEARING_TOOLS.has(toolName)) return undefined;
+  const value = input.path;
+  return typeof value === "string" && value.trim() !== "" ? value : undefined;
+}
+
+/** Expand a leading `~`, `$HOME`, or `${HOME}` in a path-denial pattern. */
+export function expandHomePattern(pattern: string): string {
+  const home = HOME.replace(/\\/g, "/");
+  if (pattern === "~" || pattern === "$HOME" || pattern === "${HOME}") {
+    return home;
+  }
+  if (pattern.startsWith("~/")) return `${home}/${pattern.slice(2)}`;
+  if (pattern.startsWith("$HOME/")) return `${home}/${pattern.slice(6)}`;
+  if (pattern.startsWith("${HOME}/")) return `${home}/${pattern.slice(8)}`;
+  return pattern;
+}
+
 export function normalizePathForMatch(path: string, cwd: string): string {
   const normalized = normalize(path);
   const rel = relative(cwd, normalized);

package/extensions/auto-mode/permissions.ts

@@ -1,5 +1,9 @@
 import type { ToolPattern } from "./types.ts";
-import { normalizePathForMatch, resolveInputPath } from "./paths.ts";
+import {
+  expandHomePattern,
+  normalizePathForMatch,
+  resolveInputPath,
+} from "./paths.ts";
 
 function normalizeToolName(name: string): string {
   const lower = name.trim().replace(/^@/, "").toLowerCase();
@@ -75,6 +79,23 @@ function getPrimaryArgument(
   return JSON.stringify(input);
 }
 
+/**
+ * Whether a resolved absolute path matches a configured path-denial pattern.
+ * Patterns support `~`/`$HOME` expansion and `*` globs, where `*` matches any
+ * characters, including `/`. Matching is case-insensitive and
+ * conservative-safe: over-matching only blocks more.
+ */
+export function matchesDeniedPath(
+  resolvedPath: string,
+  deniedPaths: string[],
+): boolean {
+  const normalized = resolvedPath.replace(/\\/g, "/");
+  return deniedPaths.some((pattern) => {
+    const expanded = expandHomePattern(pattern).replace(/\\/g, "/");
+    return wildcardToRegExp(expanded).test(normalized);
+  });
+}
+
 /** Match a scoped permission rule against a concrete tool call. */
 export function matchesToolPattern(
   pattern: ToolPattern,

package/extensions/auto-mode/state.ts

@@ -30,6 +30,7 @@ export function statusText(
   return [
     `enabled: ${(state.enabledOverride ?? config.enabled) ? "yes" : "no"}`,
     `classifier: ${config.classifierModel ?? "current session model"}`,
+    `classifier reasoning: ${config.classifierReasoningLevel ?? "server default"}`,
     `checked actions: ${state.checkedActions}`,
     `blocked actions: ${state.blockedActions}`,
     `classifier allowed: ${state.classifierAllowed}`,

package/extensions/auto-mode/types.ts

@@ -1,6 +1,34 @@
 import type { AssistantMessage } from "@earendil-works/pi-ai";
 import type { ExtensionContext } from "@earendil-works/pi-coding-agent";
 
+export type ClassifierReasoningLevel =
+  | "low"
+  | "medium"
+  | "high"
+  | "xhigh"
+  | "max";
+
+export type EffectiveClassifierReasoningLevel =
+  | "off"
+  | "minimal"
+  | ClassifierReasoningLevel;
+
+export type ClassifierReasoning =
+  | { mode: "server-default" }
+  | {
+    mode: "explicit";
+    requestedLevel: ClassifierReasoningLevel;
+    effectiveLevel: EffectiveClassifierReasoningLevel;
+  };
+
+export type ClassifierReasoningLog =
+  | ClassifierReasoning
+  | {
+    mode: "explicit";
+    requestedLevel: ClassifierReasoningLevel;
+    effectiveLevel?: undefined;
+  };
+
 /** Observability log configuration. Off by default. */
 export type LogConfig = {
   enabled: boolean;
@@ -11,6 +39,15 @@ export type LogConfig = {
 export type AutoModeSettings = {
   enabled?: boolean;
   classifierModel?: string;
+  classifierReasoningLevel?: ClassifierReasoningLevel;
+  /** When true, read-only tools (read/grep/find/ls) are classified instead of auto-allowed. */
+  classifyReadOnlyTools?: boolean;
+  /** Override the fast-stage completion token budget (default 512). */
+  fastClassifierMaxTokens?: number;
+  /** When true, file tools whose resolved path is inside the working directory are allowed deterministically (no classifier), and outside-CWD file access is classified. */
+  allowInsideWorkingDirectory?: boolean;
+  /** Path glob patterns (file tools) that are always denied before the classifier. Supports `~` and `*` (matches any characters, including `/`). */
+  deniedPaths?: unknown;
   maxUserTranscriptTokens?: number;
   maxToolTranscriptTokens?: number;
   environment?: unknown;
@@ -46,6 +83,11 @@ export type ToolPattern = {
 export type EffectiveConfig = {
   enabled: boolean;
   classifierModel?: string;
+  classifierReasoningLevel?: ClassifierReasoningLevel;
+  classifyReadOnlyTools: boolean;
+  fastClassifierMaxTokens: number;
+  allowInsideWorkingDirectory: boolean;
+  deniedPaths: string[];
   maxUserTranscriptTokens: number;
   maxToolTranscriptTokens: number;
   environment: string[];
@@ -78,12 +120,16 @@ export type DenialRecord = {
     | "permissions.deny"
     | "permissions.ask"
     | "deterministic-hard-deny"
+    | "deterministic-path-deny"
     | "classifier"
     | "setup";
 };
 
-/** Denial kind plus the read-only fast path, used for decision log entries. */
-export type DecisionKind = DenialRecord["kind"] | "read-only";
+/** Denial kind plus the deterministic allow fast paths, used for decision log entries. */
+export type DecisionKind =
+  | DenialRecord["kind"]
+  | "read-only"
+  | "inside-working-directory";
 
 export type ClassificationDecision = {
   decision: "allow" | "block";
@@ -111,6 +157,7 @@ export type ClassifierIoAttempt = {
 /** Full classifier I/O for an action, surfaced for optional observability logging. */
 export type ClassifierIo = {
   model: string;
+  reasoning: ClassifierReasoning;
   prompt: {
     system: string;
     context: string;
@@ -121,8 +168,11 @@ export type ClassifierIo = {
   durationMs: number;
 };
 
-/** Classification decision plus the I/O that produced it (when available). */
-export type ClassifyResult = ClassificationDecision & { io?: ClassifierIo };
+/** Classification decision plus resolved reasoning and the I/O that produced it (when available). */
+export type ClassifyResult = ClassificationDecision & {
+  reasoning?: ClassifierReasoningLog;
+  io?: ClassifierIo;
+};
 
 export type SettingsSources = {
   globalSettings?: SettingsFile[];

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@czottmann/pi-automode",
-  "version": "1.8.2",
+  "version": "1.10.0",
   "description": "Claude Code-style auto mode guardrail for pi.",
   "repository": {
     "url": "https://github.com/czottmann/pi-automode"
@@ -41,9 +41,9 @@
     "@earendil-works/pi-tui": "*"
   },
   "devDependencies": {
-    "@earendil-works/pi-ai": "^0.79.3",
-    "@earendil-works/pi-coding-agent": "^0.79.3",
-    "@earendil-works/pi-tui": "^0.79.3",
+    "@earendil-works/pi-ai": "^0.84.1",
+    "@earendil-works/pi-coding-agent": "^0.84.1",
+    "@earendil-works/pi-tui": "^0.84.1",
     "@types/node": "^24.0.0",
     "tsx": "^4.22.4",
     "typescript": "^5.8.0"