Enhance application features, indexing, and task management

.gitignore

@@ -27,15 +27,25 @@ legacy/tools-benchmark/tasks/eval/generated/
 legacy/tools-benchmark/tasks/eval/generated-smoke/
 legacy/tools-benchmark/results/
 
-# Solid benchmark (Phase 14 under tests/)
+# Solid benchmark (under tests/benchmark only)
 tests/benchmark/results/
+tests/benchmark/reports/
+tests/benchmark/.workspaces/
 tests/benchmark/benchmark.config.json
 tests/benchmark/fixtures/**/.mitii/
 tests/benchmark/fixtures/**/.next/
 tests/benchmark/fixtures/**/dist/
+tests/benchmark/fixtures/**/coverage/
+tests/benchmark/fixtures/**/.turbo/
+tests/benchmark/fixtures/**/.vite/
+tests/benchmark/fixtures/**/.cache/
 tests/benchmark/fixtures/**/node_modules/
 tests/benchmark/fixtures/**/package-lock.json
 tests/benchmark/fixtures/**/pnpm-lock.yaml
+tests/benchmark/fixtures/**/yarn.lock
+tests/benchmark/fixtures/**/bun.lockb
+tests/benchmark/fixtures/**/*.tsbuildinfo
+tests/reports/
 
 # Fixture sandboxes — deps, builds, Mitii session state
 legacy/tools-benchmark/fixtures/**/.mitii/
@@ -50,4 +60,10 @@ legacy/tools-benchmark/fixtures/**/docs/
 legacy/tools-benchmark/fixtures/**/index.html
 legacy/tools-benchmark/fixtures/**/src/index.css
 legacy/tools-benchmark/fixtures/**/src/main.ts
-legacy/tools-benchmark/fixtures/**/src/main.tsx
+legacy/tools-benchmark/fixtures/**/src/main.tsx
+
+# Mitii local runtime data
+.mitii-session-export.json
+.mitii-audit-pack.json
+
+.pnpm-store/

README.md

@@ -12,7 +12,7 @@
   <a href="LICENSE"><img alt="License: AGPL v3" src="https://img.shields.io/badge/License-AGPL_v3-blue.svg"></a>
   <a href="https://code.visualstudio.com/"><img alt="VS Code 1.85+" src="https://img.shields.io/badge/VS%20Code-1.85%2B-007ACC?logo=visualstudiocode"></a>
   <a href="https://nodejs.org/"><img alt="Node 20+" src="https://img.shields.io/badge/Node-20%2B-339933?logo=node.js"></a>
-  <img alt="Version 2.8.5" src="https://img.shields.io/badge/version-2.8.5-111111">
+  <img alt="Version 2.8.72" src="https://img.shields.io/badge/version-2.8.72-111111">
   <a href="https://docs.mitii.dev"><img alt="Documentation" src="https://img.shields.io/badge/docs-docs.mitii.dev-5B5BFF"></a>
 </p>
 
@@ -30,8 +30,9 @@ Mitii understands a repository before it changes it. It combines local indexing,
 
 - **Repository-aware context** — SQLite FTS5, symbols, vectors, repo maps, diagnostics, Git state, and explicitly attached files.
 - **Clear operating modes** — Ask for read-only analysis, Plan complex work, Agent applies changes, and Review inspects results.
+- **Evidence-assisted planning** — Plan mode can follow in-scope preflight diagnostics, discover first when evidence is thin, draft from the ask for scoped feature work, or ask clarifying questions when the request is too unclear.
 - **Controlled execution** — configurable approvals, dangerous-command blocking, workspace trust checks, and pre-write checkpoints.
-- **Model flexibility** — `echo` (local stub) and **OpenAI-compatible** endpoints (Ollama, LM Studio, OpenRouter, OpenAI, Azure OpenAI, DeepSeek, and similar `/v1` APIs). Native Anthropic, Gemini, and Bedrock adapters are not shipped yet.
+- **Model flexibility** — `echo` (local stub), native **Anthropic (Claude)** and **Gemini** adapters, plus **OpenAI-compatible** endpoints (Ollama, LM Studio, OpenRouter, OpenAI, Azure OpenAI, DeepSeek, and any `/v1` API).
 - **Extensible workflows** — built-in tools, MCP servers (VS Code), project rules, and reusable skills.
 - **Local evidence** — session logs and a basic audit-pack export from the VS Code host (settings redacted). Org SSO/RBAC, SIEM webhooks, and managed enterprise policy packs are not implemented yet.
 
@@ -111,18 +112,35 @@ Implement the approved plan and run the relevant tests.
 
 Mitii retrieves relevant context, selects the required capabilities, requests approval for protected actions, checkpoints affected files, applies scoped edits, and runs configured or discovered verification commands.
 
+For repair requests with matching preflight diagnostics, Mitii can skip redundant discovery and start from concrete Change steps tied to the failing files. Optional lightweight model enrichment can improve plan wording, but deterministic policy still owns gates, approvals, grants, and verification requirements.
+
+## Evidence-led execution
+
+Agent runs now carry a structured evidence artifact in addition to plan and task state. The goal is to make every major action accountable without making the live task list the source of truth.
+
+- **Discovery report** records the target, files/searches/commands inspected, bounded discovery capacity, and why discovery stopped.
+- **Issue inventory** tracks findings from diagnostics/build/verification as issues rather than just counting files.
+- **Plan evidence** records how many plan steps are linked to concrete targets, reviewed context, or verification.
+- **Execution ledger** records tool actions, edits, verification commands, and stop decisions with safe summaries and paths.
+- **Verification delta** records before/after error counts, remaining issues, checks, and the stop reason when verification is available.
+
+The plan remains the execution contract. Task lists are a derived progress view, useful for UI, but subordinate to plan evidence and verification. Once requested verification passes after edits, the agent should stop and summarize instead of continuing because the model produced transitional narration.
+
 ## CLI and SDK
 
 ### CLI
 
 ```bash
 pnpm run build:cli
+node apps/cli/bin/mitii.js --help
+node apps/cli/bin/mitii.js setup --provider ollama --yes
+node apps/cli/bin/mitii.js session
 node apps/cli/bin/mitii.js ask "Summarize the authentication flow" --echo
 node apps/cli/bin/mitii.js index --cwd /path/to/project
 node apps/cli/bin/mitii.js status --json
 ```
 
-See [apps/cli/README.md](apps/cli/README.md) for `ask`, `session`, `index`, `status`, and `export-session`. Daemon / `mitii serve` is deferred.
+New users: `mitii setup` writes non-secret provider config, then set an API key in the environment and run `mitii session` (dotted MITII banner + REPL). See [apps/cli/README.md](apps/cli/README.md) for `setup`, `ask`, `session`, `index`, `status`, and `export-session`. Daemon / `mitii serve` is deferred.
 
 ### SDK
 

apps/cli/README.md

@@ -21,6 +21,28 @@ node apps/cli/bin/mitii.js --help
 
 Legacy npm `@mitii/cli@2.7.x` is a different binary stack — prefer versions published from this tree.
 
+## First run (new users)
+
+```bash
+mitii --help                 # or: mitii -h
+mitii setup                  # pick provider + write .mitii/config.json
+export ANTHROPIC_API_KEY=…   # or GEMINI_ / OPENAI_ / MITII_API_KEY
+mitii session                # dotted MITII banner + interactive loop
+```
+
+Smoke without a live model:
+
+```bash
+mitii ask "What is recursion?" --echo
+```
+
+Check what is configured (never prints secrets):
+
+```bash
+mitii setup --show
+mitii -v                     # or: mitii --version / mitii version
+```
+
 ## Quick start
 
 ```bash
@@ -35,39 +57,164 @@ mitii export-session "Summarize this repo" --out session.json --echo
 
 | Command | Behavior |
 |---|---|
+| `setup` | Interactive (or flag-driven) model/provider setup |
 | `ask <prompt>` | SDK ask with streaming, cancel, clarify/approve |
-| `session` | Interactive prompt loop |
+| `session` | Interactive prompt loop with MITII banner |
 | `index` | Full workspace index + publish repository state |
 | `status` | Show latest persisted repository state |
 | `export-session` | Run ask and write secret-free JSON export |
-| `version` / `help` | Version and usage |
+| `version` / `help` | Version and usage (`-v` / `--version`, `-h` / `--help`) |
+
+### Modes
+
+| Mode | Behavior |
+|---|---|
+| `ask` | Q&A / explain (default) |
+| `plan` | Read-only plan; no file edits |
+| `agent` | Edit + verify with approvals |
+
+Set with `--mode <mode>` or `defaultMode` in config.
 
 ### Common options
 
 | Option | Meaning |
 |---|---|
+| `-h`, `--help` | Show usage |
+| `-v`, `--version` | Print package version |
 | `--cwd <path>` | Workspace root (default: `process.cwd()`) |
 | `--json` | Machine-readable JSON on stdout |
 | `--echo` | Force `EchoLlmPort` even when API keys are set |
 | `--clarify <text>` | Non-interactive clarification resume |
 | `--approve` / `--deny` | Non-interactive approval resume |
 | `--out <file>` | Session export path (`export-session`) |
+| `--mode <mode>` | `ask` \| `plan` \| `agent` |
+| `--loop-policy-json <json>` | Lab: one-off threshold overrides for this run |
+| `--no-loop-policy` | Ignore config `loopPolicy` for this run |
+
+Unknown options error out (they are not silently ignored).
 
 `SIGINT` cancels the active run via `run.cancel()`.
 
-## Config
+### Loop / stall policy (lab)
+
+By default the CLI uses **window-band standards** from the model context window
+(`compact` &lt; 50k, `standard` &lt; 100k, `wide` ≥ 100k). Permanent ship values live in
+`@mitii/v8` → `policy/loopPolicyBands.ts`.
+
+Optional lab overrides (same merge as VS Code Developer → Custom loop policy):
+
+```json
+{
+  "provider": "ollama",
+  "model": "qwen3-coder:30b",
+  "loopPolicy": {
+    "enabled": true,
+    "thresholds": {
+      "maxReadOnlyToolTurnsBeforeMutationNudge": 14,
+      "maxRejectedMutationRecoveries": 5
+    }
+  }
+}
+```
+
+```bash
+# One-off override (merged on top of config when enabled)
+mitii ask "Fix types" --mode agent \
+  --loop-policy-json '{"maxRejectedMutationRecoveries":5}'
+
+# Force shipped bands only for this run
+mitii ask "Fix types" --mode agent --no-loop-policy
+```
+
+Leave `loopPolicy` unset (or `"enabled": false`) for deploy / normal use.
+
+### Setup options
+
+| Option | Meaning |
+|---|---|
+| `--show` | Print current config (no secrets) |
+| `--provider <id>` | `ollama`, `anthropic`, `gemini`, `openai`, `deepseek`, … |
+| `--model <id>` | Model id |
+| `--base-url <url>` | OpenAI-compatible base URL |
+| `--global` | Write `~/.mitii/config.json` instead of project `.mitii/` |
+| `--test` | Probe the provider after writing |
+| `--yes` / `-y` | Non-interactive (requires `--provider`) |
+
+```bash
+# Local Ollama
+mitii setup --provider ollama --yes
+
+# Claude, then set the key in the shell
+mitii setup --provider anthropic --model claude-sonnet-4-5 --yes
+export ANTHROPIC_API_KEY=sk-ant-...
+
+# Custom OpenAI-compatible gateway
+mitii setup --provider openai-compatible --base-url http://localhost:1234/v1 --model local-model --yes --test
+```
+
+## Connect an API
+
+Keys go in the environment. Provider and model go in `.mitii/config.json` or `~/.mitii/config.json` (prefer `mitii setup`).
+
+```bash
+# Anthropic (Claude)
+export ANTHROPIC_API_KEY=sk-ant-...
+mitii ask "What is recursion?"
+```
+
+```json
+{ "provider": "anthropic", "model": "claude-sonnet-4-5" }
+```
 
-No secrets in files:
+```bash
+# Gemini
+export GEMINI_API_KEY=...
+# DeepSeek (OpenAI-compatible)
+export MITII_API_KEY=...
+# OpenAI
+export OPENAI_API_KEY=sk-...
+```
 
-- `.mitii/config.json` (cwd) or `~/.mitii/config.json`
-- Fields: `provider`, `model`, `baseUrl`, `workspaceId`, `defaultMode`
+```json
+{ "provider": "gemini", "model": "gemini-2.5-flash" }
+```
 
-API keys via environment only:
+```json
+{
+  "provider": "openai-compatible",
+  "providerPreset": "deepseek",
+  "model": "deepseek-chat",
+  "baseUrl": "https://api.deepseek.com/v1"
+}
+```
+
+```json
+{
+  "provider": "openai-compatible",
+  "baseUrl": "http://localhost:11434/v1",
+  "model": "qwen3-coder:30b"
+}
+```
 
-- `MITII_API_KEY` / `OPENAI_API_KEY`
-- `MITII_BASE_URL`, `MITII_MODEL` (optional overrides)
+Overrides: `MITII_PROVIDER`, `MITII_MODEL`, `MITII_BASE_URL`, `MITII_API_KEY`.
 
-Works with local OpenAI-compatible servers (Ollama, LM Studio) without a key when pointed at a local base URL.
+Local Ollama / LM Studio do not need a key. Anthropic and Gemini do.
+
+Cursor Cloud Agents are a separate agent API, not an LLM endpoint. Point `openai-compatible` at any `/v1/chat/completions` proxy if you need a custom gateway.
+
+## Session UI
+
+`mitii session` prints a dotted **MITII** banner, then workspace / provider / mode, and the `mitii>` prompt. If you are still on the echo stub, it points you at `mitii setup`.
+
+## Troubleshooting
+
+| Symptom | What to try |
+|---|---|
+| Echo / stub answers only | `mitii setup`, then export the matching API key |
+| `unknown option` | Typos fail loudly — run `mitii --help` |
+| Index falls back to snapshot | Optional native deps / embeddings; ask still works with host snapshot |
+| No repository state | `mitii index`, or let `ask` auto-index |
+| Wrong model | `mitii setup --show`, then `mitii setup` again |
 
 ## Out of scope
 
@@ -80,6 +227,7 @@ pnpm --filter @mitii/cli typecheck
 pnpm --filter @mitii/cli test
 pnpm --filter @mitii/cli build
 node apps/cli/bin/mitii.js ask "ping" --echo --json
+node apps/cli/bin/mitii.js setup --show
 ```
 
 ## Links

apps/cli/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@mitii/cli",
-  "version": "2.8.5",
+  "version": "2.8.72",
   "description": "Mitii headless CLI over @mitii/sdk.",
   "license": "AGPL-3.0-or-later",
   "publishConfig": {
@@ -40,6 +40,8 @@
     "vitest": "^3.2.7"
   },
   "optionalDependencies": {
-    "@lancedb/lancedb": "0.33.0"
+    "@lancedb/lancedb": "0.33.0",
+    "onnxruntime-node": "1.21.0",
+    "onnxruntime-web": "1.21.0"
   }
 }

apps/cli/src/banner.ts

@@ -0,0 +1,56 @@
+/**
+ * Terminal session chrome for the Mitii CLI.
+ * Prefer ASCII/Unicode that survives common terminal fonts.
+ */
+
+export const MITII_BANNER = `
+  ·   · ····· ····· ····· ·····
+  ·· ··   ·     ·     ·     ·
+  · · ·   ·     ·     ·     ·
+  ·   ·   ·     ·     ·     ·
+  ·   · ·····   ·   ····· ·····
+`.replace(/^\n|\n$/g, '');
+
+export interface SessionHeaderOptions {
+  cwd: string;
+  providerLabel: string;
+  mode: string;
+  version: string;
+  /** True when running the local echo stub (no live model). */
+  isEcho?: boolean;
+  /** Point new users at `mitii setup` when no provider is configured. */
+  showSetupHint?: boolean;
+}
+
+export function formatSessionHeader(options: SessionHeaderOptions): string {
+  const lines = [
+    MITII_BANNER,
+    '',
+    `  Mitii CLI v${options.version}  ·  headless agent`,
+    `  workspace  ${options.cwd}`,
+    `  provider   ${options.providerLabel}`,
+    `  mode       ${options.mode}`,
+    '',
+  ];
+
+  if (options.showSetupHint) {
+    lines.push(
+      '  No live model yet — answers use the local echo stub.',
+      '  Run  mitii setup  to choose a provider and write config.',
+      '',
+    );
+  } else if (options.isEcho) {
+    lines.push(
+      '  Echo mode (--echo or provider=echo) — local stub, no remote API.',
+      '',
+    );
+  }
+
+  lines.push(
+    '  Empty line or Ctrl-D to exit · Ctrl-C cancels the active run',
+    '  Modes: ask (Q&A) · plan (read-only plan) · agent (edit + verify)',
+    '',
+  );
+
+  return `${lines.join('\n')}\n`;
+}