devcouncil 0.2.0 → 0.2.1

package/README.md

@@ -1,346 +1,284 @@
-# DevCouncil: The Gated AI Orchestrator
+# DevCouncil: Components and Modules for AI Development
+
+[![Website](https://img.shields.io/badge/website-devcouncil.vbcr.dev-10B981?style=flat&logo=safari&logoColor=white)](https://devcouncil.vbcr.dev/)
+[![CI](https://github.com/bharathvbcr/DevCouncil/actions/workflows/ci.yml/badge.svg)](https://github.com/bharathvbcr/DevCouncil/actions/workflows/ci.yml)
+[![npm version](https://img.shields.io/npm/v/devcouncil)](https://www.npmjs.com/package/devcouncil)
+[![License](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)
 
 <p align="center">
-  <img src="https://raw.githubusercontent.com/bharathvbcr/DevCouncil/main/src/devcouncil/assets/devcouncil_logo_premium.png" alt="DevCouncil Logo" width="300">
+  <a href="https://devcouncil.vbcr.dev/"><strong>Explore the Interactive Architecture &amp; Code Graph Showcase (devcouncil.vbcr.dev) &rarr;</strong></a>
 </p>
 
-[![License](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)
-[![Python 3.12+](https://img.shields.io/badge/python-3.12+-blue.svg)](https://www.python.org/downloads/)
-[![uv](https://img.shields.io/badge/managed%20by-uv-purple.svg)](https://github.com/astral-sh/uv)
+> **"DevCouncil should not merely generate code. It should make AI-generated work prove that it satisfied the original intent."**  
+> *Evidence, not model confidence, is the final authority.*
 
-**"DevCouncil should not merely generate code. It should make AI-generated work prove that it satisfied the original intent."**
+DevCouncil is **components and modules** for AI-assisted software development. It ships independently installable binaries and libraries — code intelligence (`devmap`), task and lease state (`dcstore`), deterministic verification (`dcverify`), ignore-aware search (`dcgrep`), and a Go host (`devcouncil` / `dev`) — so a harness or an app can take only the pieces it needs and update them one at a time.
 
-DevCouncil is a high-integrity command-line orchestration platform for AI-assisted software development. It turns AI implementation from a black-box generation task into a gated engineering workflow where every change is authorized, verified, and traceable back to a requirement.
+See [repository input and verification boundaries](docs/SECURITY_BOUNDARIES.md)
+for executable selection, filesystem protections, resource limits, and the
+difference between unavailable evidence and a passing check.
 
-DevCouncil does not replace coding agents. It sits beside tools like Codex CLI, Gemini CLI, Claude Code, OpenCode, Google Antigravity CLI, Warp/Oz, Cursor, Aider, and bring-your-own prompt-taking CLIs, then owns the plan, task scope, verification loop, repair prompts, and evidence trail.
+**Manvi wraps these components** into a coding-agent harness (turn loop, providers, policy, TUI, `manvi serve`). **GitPulse** uses Manvi for policy, workbench, and agent hosting, and DevCouncil components for code intelligence and related analysis. DevCouncil does not replace coding agents; it sits beside Claude Code, Codex, Cursor, and others as selectable modules.
 
-## Documentation
+---
 
-- [Quickstart](docs/quickstart.md): shortest install-to-first-task path.
-- [Daily workflow](docs/workflow.md): manual sidecar loop, verification, repair, and rollback.
-- [Coding CLI integration](docs/coding-cli-integration.md): Codex, Gemini, Claude Code, OpenCode, Antigravity, Cursor, Aider, MCP, hooks, and automated executors.
-- [Integration tiers](docs/integration-tiers.md): headless executor vs MCP-only vs sidecar definitions.
-- [CLI command reference](docs/cli-reference.md): available `dev` commands.
-- [Architecture](docs/architecture.md): components, artifact graph, state machine, and gated execution.
-- [Executor adapters](docs/executor-adapters.md): manual, coding CLI, native-preview, Mini-SWE, and OpenHands execution paths.
-- [Live review](docs/live-review.md): `dev watch` session review, cards, signals, and blocking behavior.
-- [Model routing](docs/model-routing.md): provider selection, role models, OpenRouter, Vertex AI, Doubleword, and Ollama (local) setup.
-- [Security model](docs/security.md): redaction, permissions, allowlists, and local state.
-- [Project status](docs/project-status.md): current maturity by subsystem.
-- [Roadmap](docs/roadmap.md): planned work.
+## Architecture & Components
 
-## Why DevCouncil Exists
+DevCouncil's core runtime is a set of standalone, compiled native modules. Each binary is useful on its own, sits on `PATH`, and can be spawned or linked by a host that wants that one job:
 
-Standard AI coding agents are good at producing the happy path, but they often fail in expensive ways when complexity grows:
+| Binary | Language | Role & Ownership |
+|--------|----------|------------------|
+| `devcouncil` / `dev` | **Go** | Host orchestrator: Stdio MCP server (`mcp`), multi-host agent integration (`integrate`), engineering skills distribution (`skills`), task verification (`verify`), and devmap forwarding (`map`, `graph`, `ast`). |
+| `devmap` | **Rust** | Code intelligence engine: 36+ tree-sitter language extractors, symbol resolution, blast-radius calculation, dead-code analysis, workspace guides (`AGENTS.md`), and DevMap MCP server (`devmap mcp`). |
+| `dcstore` | **Rust** | State & lease store: SQLite-backed task repository, mutual-exclusion leases for concurrent agent building, evidence records, and gap tracking. |
+| `dcverify` | **Rust** | Deterministic verification: Unified-diff parsing, planned file scope classification, anti-laziness/stub gates, test coverage evaluation, and typed `next_actions` repair signals. |
+| `dcgrep` | **Rust** | Code search: Ripgrep-powered ignore-aware search engine with optional trigram indexing (`tgrep-core`). |
 
-- **Requirement omission:** agents lose track of original product or PRD constraints across chat turns.
-- **Architecture drift:** agents add dependencies or change design patterns without explicit authorization.
-- **Unverified success:** agents claim tests passed without proving that the new logic was exercised.
-- **Hidden assumptions:** important decisions stay buried in transient chat history instead of durable project artifacts.
+### How hosts consume the modules
 
-**DevCouncil makes evidence, not model confidence, the final authority.**
+- **Manvi** wraps the components: it drives the agent loop, provider routing, policy ladder, and TUI, and reaches each binary over JSON on stdio. A host that wants a harness embeds Manvi rather than reimplementing it.
+- **GitPulse** uses both for their respective jobs: Manvi (`manvi serve`) for policy, workbench, and agent hosting; selected DevCouncil crates and the `devmap` CLI for in-process code intelligence. It does not have to take every module.
+- **Other apps** pick the same way — `devmap` alone, `dcverify` alone, the Go host MCP, or the full set. Updating one module does not require shipping the rest.
 
-It creates a persistent **Requirement -> Task -> Diff -> Evidence** graph, blocks completion when evidence is missing, detects unauthorized changes, and produces a final report that can be reviewed like an engineering artifact.
+> [!NOTE]
+> DevCouncil is built entirely as native compiled Go and Rust binaries; legacy Python orchestration has been retired (see [docs/PHASE7_LONG_TAIL.md](docs/PHASE7_LONG_TAIL.md)).
 
-## Quickstart
+---
 
-Run DevCouncil commands in a normal terminal from the root of the repository you want DevCouncil to manage. Do not run these commands inside a coding CLI chat.
+## Installation
 
-Install `uv` first if it is missing:
+DevCouncil supports macOS, Linux, and Windows. Requires a Go toolchain (`>=1.22`), Rust/`cargo`, and Git.
 
-```powershell
-powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
-```
+### 1. Build and Install Native Binaries
 
-On macOS or Linux:
+From a clone of this repository:
 
 ```bash
-curl -LsSf https://astral.sh/uv/install.sh | sh
-```
+# macOS & Linux: builds Go host and all Rust analysis binaries into ~/.local/bin
+bash scripts/install.sh
 
-Install DevCouncil from npm:
+# Standalone DevMap (no Go host)
+bash scripts/install.sh --only=devmap
 
-```bash
-npm install -g devcouncil
-devcouncil --help
-dev --help
+# Analysis suite only (devmap dcstore dcverify dcgrep)
+bash scripts/install.sh analysis
+
+# Windows (PowerShell):
+.\scripts\install.ps1
+.\scripts\install.ps1 -Components devmap
 ```
 
-Start the first gated workflow from your target repository:
+To build and install the analysis components (`devmap`, `dcstore`, `dcverify`, `dcgrep`):
 
 ```bash
-cd path/to/your/project
-dev setup
-dev plan "Describe the implementation goal"
-dev tasks
-dev run TASK-001 --executor manual
-dev prompt TASK-001
-dev verify TASK-001
-```
+# Build & install all analysis components to ~/.local/bin
+bash scripts/install-components.sh
 
-On a fresh interactive setup, DevCouncil can configure supported coding CLI integrations immediately; pass `--skip-integrations` if you want to defer that step.
+# Or install specific components
+bash scripts/install-components.sh devmap
+bash scripts/install-components.sh dcstore dcverify dcgrep
 
-### Run locally on macOS (Apple Silicon + Ollama)
+# Build the Go host binary
+go -C backend/go_orchestrator build -o ~/.local/bin/devcouncil ./cmd/devcouncil
+ln -sf devcouncil ~/.local/bin/dev
+```
 
-DevCouncil runs fully offline against [Ollama](https://ollama.com) — no API key, no per-token cost. It is Apple-Silicon-aware: `dev setup --provider ollama` sizes the default local model to your Mac's unified memory, and `dev doctor` reports the chip/RAM, pings the Ollama server, and flags a too-small context window.
+Ensure `~/.local/bin` is in your `PATH`:
 
 ```bash
-brew install ollama && ollama serve
-ollama pull qwen2.5-coder:32b     # use the size `dev doctor` recommends for your RAM
-export OLLAMA_NUM_CTX=16384       # large planning prompts need a raised context window
-dev setup --provider ollama      # auto-selects the model for your RAM
+export PATH="$HOME/.local/bin:$PATH"
 ```
 
-See [Model routing → macOS / Apple Silicon](docs/model-routing.md) for the RAM-to-model table.
-
-Paste only the output from `dev prompt TASK-001` into Codex, Gemini, Claude Code, OpenCode, Antigravity, Warp, Cursor, Aider, or another coding tool. Keep `dev setup`, `dev plan`, `dev run`, and `dev verify` in the terminal at the repository root.
+### 2. Optional Global npm Shim
 
-For an automated end-to-end run with a supported coding CLI installed:
+If you prefer to dispatch through npm, install the lightweight Node.js wrapper (Node 18+). The npm package ships a shim that resolves and executes the native binaries:
 
 ```bash
-dev e2e "Describe the implementation goal" --executor codex
-dev e2e "Describe the implementation goal" --executor antigravity
-dev e2e "Describe the implementation goal" --executor warp
-dev go "Describe the implementation goal" --executor codex
+npm install -g devcouncil
 ```
 
-`dev e2e` is the explicit one-command integration target for coding agents. It initializes local DevCouncil state if needed, plans the goal, runs each approved task through the selected executor, verifies the resulting diff, and prints the final report. If `--executor` is omitted, DevCouncil uses `execution.default_executor` from `.devcouncil/config.yaml`. `dev go` is kept as a shorter alias for the same flow.
-
-For machine-readable agent handoff, write the final report to a stable file:
+### 3. Verify Environment
 
 ```bash
-dev e2e "Describe the implementation goal" --executor codex --agent
-dev e2e "Describe the implementation goal" --executor codex --json --report-file .devcouncil/reports/latest.json
+devcouncil --help
+dev --help
+devmap --version
+dcverify health
+dcstore --db .devcouncil/state.sqlite health
+dcgrep health
 ```
 
-`--agent` enables JSON output and writes `.devcouncil/reports/latest.json`. Fresh projects default to manual sidecar mode, so pass an automated executor or set `execution.default_executor` before using `dev e2e` without `--executor`.
-
-See the full [quickstart](docs/quickstart.md) for installation variants, API-key setup, and first-run guidance.
+---
 
-OpenCode and Google Antigravity CLI are built-in executors and MCP integrations:
+## Core Capabilities & Workflows
 
-```bash
-dev integrate opencode --apply
-dev run TASK-001 --executor opencode
-dev agents run TASK-001 --agent opencode --profile default
-dev integrate antigravity --apply
-dev run TASK-001 --executor antigravity
-dev agents run TASK-001 --agent agy --profile default
-```
+### 1. Coding Agent Integration (`devcouncil integrate`)
 
-Register any other local CLI that accepts prompts. `dev agents` is the first-class agent hub; `dev integrate cli-agent` remains available for older scripts:
+Connect DevCouncil's MCP server to Cursor or Claude Code. `--write-gate` has been removed: it named a pre-tool-use gate that only the retired lifecycle hooks installed, and nothing enforced it. Use `dev hook status` to inspect old registrations and `dev hook disable --dry-run` to preview cleanup.
 
 ```bash
-dev agents add myagent --command myagent --arg run --input-mode prompt-file --prompt-arg=--prompt-file --supports-mcp
-dev agents
-dev agents doctor
-dev agents run TASK-001 --agent myagent --profile default
-```
+devcouncil integrate cursor --apply
+devcouncil integrate claude --apply
 
-GEPA prompt-profile optimization is available for the agent hub:
+# Stub receipt only — not a working installer:
+# devcouncil integrate antigravity --apply
 
-```bash
-dev agents optimize --agent codex --profile yolo --evals .devcouncil/evals/agent-profile.jsonl --dry-run
-dev agents optimize --agent codex --profile yolo --evals .devcouncil/evals/agent-profile.jsonl --apply
+# Verify existing integration configuration
+devcouncil integrate cursor --check
 ```
 
-## Feature Set
-
-DevCouncil is an application layer around coding agents. It does not just emit prompts; it owns the workflow state, validates task scope, records evidence, and produces release-style reports.
+### 2. Engineering Skills Delivery (`devcouncil skills`)
 
-### Workflow Features
+Deliver verified engineering practices and code-intelligence skills directly into agent skill folders (`.agents/skills`, `.claude/skills`, `.cursor/skills`):
 
-- **Repository onboarding:** `dev setup` initializes `.devcouncil/`, generates the repo map + `AGENTS.md`/`CLAUDE.md` guides, scaffolds applicable engineering skills, runs environment checks, offers integration setup, and prints the next useful commands. Use `--skip-map` / `--skip-skills` to opt out, or `--scaffold-ci` to also write a starter GitHub Actions workflow.
-- **Repository mapping:** `dev map` writes `.devcouncil/repo_map.json`, identifies important files and subsystems, filters generated/temp files, and keeps managed `AGENTS.md` / `CLAUDE.md` workspace guides synchronized. Subsystems, entry points, neighbors, and important surfaces are now inferred generically for **any** repository — grouped from the directory tree and ranked by an import-graph in-degree — so the map (and the structural context it feeds into prompts) is meaningful outside DevCouncil's own tree, not just within it. The map records the git HEAD and tracked-file fingerprint it was built from; when prompts reuse a map that has fallen behind the current code, they flag it as stale (run `dev map` to refresh) rather than silently feeding wrong structure. The map is also generated automatically on first init.
-- **Engineering skills:** `dev skills` lists the bundled skills and shows which apply to the repository; `dev skills scaffold` writes them into `.claude/skills/<name>/SKILL.md`. A merged always-on `core-engineering` skill (think-before-coding, simplicity, surgical changes, goal-driven execution, evidence-grounded communication) plus domain skills (Android, iOS, Windows, web, AI training) that brief the agent on current SDKs, deprecations, and tooling before coding. Applicable skills are also embedded into `dev prompt` output.
-- **CI scaffolding:** `dev scaffold-ci` writes a starter `.github/workflows/devcouncil.yml` derived from the configured test/lint/typecheck commands, filtered to the detected language stack; it never overwrites existing CI unless `--force`.
-- **Planning council:** `dev plan` turns a goal into requirements, acceptance criteria, assumptions, critique findings, and executable tasks.
-- **Task graph:** `dev tasks` and `dev show TASK-001` expose requirement links, acceptance-criterion links, planned files, expected tests, allowed commands, forbidden changes, dependencies, and status. Tasks can declare `depends_on`; the plan gate rejects unknown dependencies and cycles, and `dev go`/`dev e2e` run tasks in topological order and skip a task whose prerequisites didn't complete (rather than letting it fail spuriously and burn its repair budget).
-- **Scoped task prompts:** `dev prompt TASK-001` creates a constrained implementation prompt for sidecar agents, including file scope, verification expectations, and forbidden changes. The prompt now embeds the current (secret-redacted) contents of each planned file with a top-level symbol outline, structural orientation (from the code-review graph when available, otherwise the generated `repo_map.json`), and a **dependents (blast-radius) list** — the files that import each file being changed, from the map's precomputed reverse-import index — so the agent edits in place and keeps call sites working instead of starting blind. A central prompt budget keeps the core (goal/scope/instructions) always present and fits the optional context sections in priority order (file contents > structural > dependents > skills), dropping the lowest-priority ones with an explicit marker rather than overflowing silently.
-- **Execution:** `dev run TASK-001` supports manual sidecar mode, built-in coding CLI executors, external executors, and registered custom CLI agents.
-- **One-command flow:** `dev e2e "goal"` and `dev go "goal"` can initialize state, plan, run approved tasks, verify the diff, and generate a report. With an automated executor the run is now a **closed loop**: a task that fails verification is re-driven through a bounded self-repair loop (a correction manifest is written and the executor re-run) until it verifies or the `execution.max_repair_attempts` budget is spent, with no-progress detection that stops early when the same blocking gaps reappear.
-- **Verification:** `dev verify TASK-001` captures the diff, runs expected evidence commands, checks planned-file compliance, detects orphan changes, flags unplanned dependency edits, scans for secrets, and links evidence to acceptance criteria. An **empty diff can no longer pass** a task that declares files to create or modify (work that committed earlier is still recognized via the task checkpoint), and the result reports the rigor it ran at (`verification_mode` compiled vs coarse, `diff_empty`, `coverage_measured`/`coverage_skipped_reason`) plus a distinct `advisory_actions` list so an agent never mistakes "passed" for "proven." `dev verify` exits non-zero when a task is blocked so shell-driven agents can gate on `$?`.
-- **Repair:** `dev repair` converts blocking gaps into focused follow-up work instead of leaving failures as vague test output.
-- **Rollback:** `dev rollback TASK-001` uses task checkpoints to revert scoped work when a task needs to be backed out.
-- **Reporting:** `dev report` emits a requirements coverage table, evidence summary, blocking gaps, and live-review blockers; JSON and PR-comment paths are available for automation.
+```bash
+# List available skills embedded in the binary
+devcouncil skills list
 
-### App Surfaces
+# Scaffold all applicable skills into the repository
+devcouncil skills scaffold
 
-- **CLI:** `dev` and `devcouncil` expose the same Typer command surface for local terminal workflows.
-- **Agent hub:** `dev agents` lists built-in and custom agents, `dev agents add` registers prompt-taking CLIs, `dev agents doctor` checks wiring, `dev agents run` executes a task through a named agent/profile, and `dev agents optimize` uses GEPA to tune profile preambles from offline eval examples.
-- **Integration hub:** `dev integrate all --apply` configures supported coding CLI and MCP integrations; targeted setup exists for Codex, Gemini, Claude Code, OpenCode, Antigravity, Cursor, Warp/Oz, hooks, and custom CLI agents. `dev integrate check` now reports each client's **enforcement posture** — `pre-action` (a native hook blocks unauthorized writes before they happen) vs `verify-only` (forbidden changes are caught only after the fact by verification) — so the containment guarantee isn't overstated for clients without a pre-action gate.
-- **MCP server:** `dev mcp-server` exposes DevCouncil context and workflow tools over stdio for MCP-capable clients. `devcouncil_verify_task` now runs DevCouncil's strong compiled per-criterion checks when a provider key is configured (falling back to a clearly-labeled `coarse` mode otherwise), refuses to pass on an empty diff, and returns `verification_mode`, `diff_empty`, `coverage_measured`/`coverage_skipped_reason`, and an `advisory_actions` array alongside the blocking `next_actions`. Cheap, re-verify-free read tools — `devcouncil_get_gaps` and `devcouncil_get_next_actions` — let a reconnecting agent resume outstanding work from persisted gaps (which now carry `file`/`line`/`suggested_command`/`acceptance_criterion_id`). Task leases expire on a config-driven TTL so a crashed agent's task frees itself, with `devcouncil_renew_lease` and `devcouncil_list_leases` for long runs and fleet supervision; a partial-unique DB index enforces a single active lease per task, so concurrent checkouts can't both win the writer slot. A pure-MCP agent can now make the change itself through lease-gated write tools — `devcouncil_write_file` and `devcouncil_apply_patch` — which policy-check every target path *before* it lands (out-of-scope, protected, or escaping paths are rejected; a patch with any out-of-scope target is rejected whole, never partially applied), write atomically, and record a `FileChangeEvent` for provenance. The corpus is also browsable as MCP **resources** (`devcouncil://report`, `devcouncil://tasks`, `devcouncil://gaps`, `devcouncil://cards`, `devcouncil://task/{id}`) so a host can read project state without a tool call. `devcouncil_get_task_provenance` then exposes that audit trail — gated file changes, verification runs, diff-coverage evidence, and the latest correction manifest — so what happened on disk is inspectable. The diff↔coverage proof is now also retained across graph reloads (it was previously dropped), so reports and `dev status` reflect whether the changed lines were actually exercised.
-- **Live review:** `dev watch` tracks review cards, signals, blocking feedback, and repair guidance while a session is active.
-- **Trace viewer:** `dev trace tail --follow` streams local DevCouncil trace events for execution, verification, and agent handoff.
-- **Dashboard:** `dev dashboard --open` serves a local status dashboard and opens it in the default browser for project state and live workflow visibility.
-- **Agent-consumable CLI:** machine output for shell-driven agents — `dev prompt --json` (`{ok, task_id, prompt}`), `dev handoff --json` (`{ok, manifest_path, run_id, next_command}` to chain `dev run`), `dev verify` exits non-zero when blocked, and `dev status`/`dev report` accept `--fail-on-blocking` to exit non-zero on outstanding blocking gaps so a loop can gate on `$?`.
-- **Config editor:** `dev config` and `dev config models` inspect/update provider, model, executor, and command configuration.
-- **Artifact tools:** `dev artifacts validate` checks stored graph integrity.
-- **Code intelligence:** `dev lsp inspect` checks optional language-server readiness, and `dev ast match` searches code structurally.
-- **Doctor:** `dev doctor` validates local dependencies, commands, and environment prerequisites before a workflow fails deeper in execution.
+# Scaffold a specific skill (e.g. core-engineering or devmap)
+devcouncil skills scaffold --skill core-engineering
+devcouncil skills scaffold --skill devmap
+```
 
-### Agent And Executor Support
+### 3. Repository Mapping & Code Intelligence (`devmap` / `dev map`)
 
-DevCouncil works with human-in-the-loop sidecar sessions and automated prompt handoff:
+Build and query deep semantic relationships across your codebase without an LLM:
 
-- **Manual sidecar:** paste `dev prompt TASK-001` into any agent, then run `dev verify TASK-001`.
-- **Built-in coding CLI adapters:** `codex`, `gemini`, `claude`, `opencode`, `antigravity`, `warp`, `cursor`, `aider`, and aliases such as `codex-cli`, `gemini-cli`, `claude-code`, `opencode-cli`, `antigravity-cli`, `agy`, `agy-cli`, `warp-cli`, `oz`, `cursor-agent`, and `cursor-cli`.
-- **Custom CLI agents:** register any prompt-taking command with stdin, argument, or prompt-file handoff.
-- **Execution profiles:** custom agents can use profiles such as `default`, `yolo`, and `prod` to adjust prompt constraints while DevCouncil still verifies the final diff.
-- **External automated adapters:** `mini`, `openhands`, `native-preview`, and `native` are available when the corresponding local executor is configured.
-- **Hook-aware clients:** `dev integrate hooks --apply` installs write/shell hooks for Codex, Gemini, Claude, Cursor, and OpenCode so DevCouncil policy can block unauthorized actions before verification. The post-task hook can run deterministic verification of the active task and record gaps (enable `execution.verify_on_post_task`; off by default to keep hooks fast). File-write policy uses one shared path normalizer across the hook and task-policy paths that resolves every target and **denies anything outside the project root** (so the path that's checked is the path that's enforced), and the pre-tool-use hook is fail-closed: an unparseable or error payload is surfaced (and blocked under `--strict`/`DEVCOUNCIL_HOOK_STRICT`) rather than silently allowed.
+```bash
+# Build the repository map (.devcouncil/repo_map.json) and code graph
+devmap build --manifest
 
-### Gates And Evidence
+# Shorthand via the Go host:
+dev map
 
-DevCouncil blocks completion on concrete gaps rather than model confidence:
+# Query symbol blast radius & reverse dependents before editing
+devmap impact path/to/file.go
 
-- **Plan approval gates:** requirements must have acceptance criteria, acceptance criteria need verification methods, tasks must map to known requirements and acceptance criteria, high-impact assumptions must be resolved, and high/critical critique findings must be closed.
-- **Task readiness gates:** the working tree must be clean for the task, planned files must be declared, and each task needs allowed commands plus expected verification evidence.
-- **Diff gates:** verification detects files changed outside the planned task scope, dependency-file edits made without authorization, deleted/added files, and untracked file diffs.
-- **Evidence gates:** passing evidence commands are linked back to acceptance criteria; missing passing evidence becomes a blocking gap.
-- **Security gates:** secret scanning runs over captured diffs, and command output is redacted before it is written to logs.
-- **Live-review gates:** unresolved critical review cards can block task verification and appear in reports.
+# Trace dependency paths between two symbols
+devmap trace SymbolA SymbolB
 
-### Providers, Models, And Cost Tracking
+# Find dead code with confidence tiers (extracted | inferred | ambiguous)
+devmap dead
 
-- **Providers:** OpenRouter, Vertex AI, Doubleword, and Ollama (local, no key) are supported through local configuration and secrets.
-- **Role models:** planner, critic, arbiter, reviewer, and repair roles can share one model or use per-role overrides.
-- **Structured repair:** model routing includes JSON repair paths for structured planning and review outputs.
-- **Model defaults:** packaged YAML defaults ship with the tool so installed CLI environments do not depend on source-tree-only files.
-- **Telemetry:** local trace and cost data feed `dev status`, reports, and dashboard surfaces.
+# Launch the interactive HTML visualizer
+devmap view
+```
 
-### Reports And Automation Outputs
+### 4. Task Leases & Gated Verification (`devcouncil verify`)
 
-- **Markdown reports:** include verdict, coverage summary, requirement/task mapping, blocking gaps, and live-review status.
-- **JSON reports:** `--json` and `--report-file` support machine-readable handoff to other automation.
-- **Agent preset:** `--agent` writes `.devcouncil/reports/latest.json` for stable downstream consumption.
-- **PR comments:** `dev report --github-pr-comment` and `dev report --gitlab-pr-comment` can publish verification summaries to pull/merge requests.
-- **GitHub checks:** preview GitHub report/check surfaces are available for repository automation.
+Prevent multi-agent file trampling and verify diff correctness against deterministic criteria:
 
-### Local State And Files
+```bash
+# Run deterministic verification against a task's diff
+devcouncil verify TASK-001
 
-DevCouncil stores local workflow state in the target repository:
+# Machine-readable output for agent loops
+devcouncil verify TASK-001 --json
 
-- `.devcouncil/config.yaml`: provider, executor, command, integration, and workflow settings.
-- `.devcouncil/secrets.env`: local provider secrets such as API keys or Vertex AI project/location values. Git-ignored; copy `.devcouncil/secrets.env.example` and fill in real values. Environment variables take precedence over this file.
-- `.devcouncil/repo_map.json`: generated repository map and subsystem navigation index.
-- `.devcouncil/state.sqlite`: SQLite state for requirements, assumptions, tasks, evidence, gaps, critique findings, and project phase history.
-- `.devcouncil/checkpoints/`: task snapshots used by verification and rollback.
-- `.devcouncil/logs/`: redacted stdout/stderr from verification commands.
-- `.devcouncil/runs/<run-id>/agent-run.json`: prompt, executor, profile, exit status, and run metadata for automated agent executions.
-- `.devcouncil/reports/latest.json`: optional machine-readable report generated by `dev e2e --agent`.
-- `.devcouncil/integrations/` and `.agents/`: generated integration files such as Warp/Oz MCP JSON and Antigravity MCP config.
+# Sandbox flag is recorded only; docker/nix do not isolate (TASK-P7-2)
+devcouncil verify TASK-001 --sandbox local
+```
 
-### Maturity
+Verification today (Go `verify.Run()`) enforces:
+- **Planned file scope:** Rejects unauthorized edits outside declared task boundaries.
+- **Orphan diffs / dependency-risk:** Flags files changed that were not planned.
+- **Expected tests / allowed commands:** Runs the task's command list via `/bin/sh -c` in the project root.
+- **Typed `next_actions`:** Structured repair instructions when those gates fail.
 
-The stable daily workflow is planning, manual sidecar execution, verification, repair, rollback, and reporting. Coding CLI executors, MCP, live review, dashboard, PR comments, LSP/AST tools, and GitHub check surfaces are preview features. Native autonomous execution is experimental and still requires DevCouncil verification before work is considered complete.
+Stub detection, secret scanning, and coverage intersection live in the **`dcverify`** binary. `devcouncil verify`, MCP `devcouncil_verify_task`, and **Manvi** `runRigor` all spawn it. When `dcverify` is not installed the report carries an empty `rigor_applied` **and** a `rigor_skipped_reason` naming what to install — never a silent clean pass. Diff↔coverage runs only when a profile is supplied (`devcouncil verify --coverage PATH`); otherwise `coverage_skipped_reason` says so.
 
-## Core Flow
+### 5. MCP Servers for AI Agents
 
-DevCouncil's recommended default is **Manual Sidecar Mode**:
+DevCouncil provides two complementary Model Context Protocol (MCP) servers:
 
-1. DevCouncil plans the work and creates a task graph.
-2. You ask DevCouncil for one constrained task prompt.
-3. You paste that prompt into your coding CLI or agent.
-4. The agent edits the repository.
-5. DevCouncil verifies the resulting diff against task constraints.
-6. If verification fails, DevCouncil creates a focused repair loop.
+- **Host MCP Server (`devcouncil mcp`)**: Eight tools — checkout / renew / release / next_task / get_diff / verify_task / get_gaps / policy_check_write. `verify_task` runs Go `verify.Run()`, which spawns `dcverify` for the stub and secret gates; diff↔coverage needs a profile this tool has no field for, so use `devcouncil verify --coverage PATH`.
+- **DevMap MCP Server (`devmap mcp`)**: Eleven query tools (status, search, dependencies, impact, trace, neighbors, dead_symbols, clones, preview, explore, affected_tests). Always pass `repo_path`.
 
-The detailed task-by-task workflow lives in [docs/workflow.md](docs/workflow.md).
+---
 
-## How The Repo Runs
+## Architecture Flow
 
 ```mermaid
 flowchart TD
-    user["User runs dev/devcouncil"] --> cli["Typer CLI\nsrc/devcouncil/cli/main.py"]
-    cli --> config["Config + secrets\n.devcouncil/config.yaml\n.devcouncil/secrets.env"]
-    cli --> map["Repo map\nsrc/devcouncil/indexing/repo_mapper.py"]
-    cli --> planning["Planning commands\ndev plan / dev prompt / dev tasks"]
-
-    config --> providers["Model providers\nOpenRouter, Vertex AI, Doubleword, or Ollama"]
-    providers --> router["ModelRouter\nrole models, cache, telemetry, structured JSON repair"]
-    router --> planning
-
-    planning --> storage["SQLite + repositories\nrequirements, tasks, gaps, evidence, state"]
-    storage --> artifactGraph["Artifact graph\nRequirement -> Task -> Diff -> Evidence"]
-    artifactGraph --> gates["Gate policy\nplanned files, commands, secret checks"]
-
-    gates --> manual["Manual sidecar\ndev prompt + user agent edits"]
-    gates --> coding["Coding CLI executor\nCodex, Gemini, Claude, OpenCode, Antigravity, Warp, custom CLIs"]
-    gates --> native["Native preview executor\nLLM router + TaskRunner"]
-    gates --> external["Mini-SWE / OpenHands adapters"]
-
-    coding --> runlog["Run artifacts\nprompt file, redacted logs, manifest, trace events"]
-    native --> runlog
-    external --> runlog
-    manual --> diff["Repository diff"]
-    runlog --> diff
-
-    diff --> verify["Verifier\ndev verify / automatic post-run verification"]
-    verify --> evidence["Evidence + gaps"]
-    evidence --> storage
-    evidence --> repair["Repair loop\ndev repair / dev watch repair"]
-    evidence --> report["Reports\ndev report, JSON, GitHub/GitLab comments"]
-
-    cli --> mcp["MCP server\ndev mcp-server"]
-    mcp --> storage
-    mcp --> artifactGraph
-    mcp --> repair
-
-    cli --> live["Live review\ndev watch"]
-    live --> cards["Cards + signals\nblocking review feedback"]
-    cards --> report
+    subgraph Agents["Harnesses, hosts, and coding agents"]
+        Claude["Claude Code"]
+        Cursor["Cursor"]
+        Codex["Codex"]
+        AGY["Antigravity"]
+        Warp["Warp"]
+        Manvi["Manvi (wraps these components)"]
+        GitPulse["GitPulse (selects modules)"]
+    end
+
+    subgraph Host["Go Host Orchestrator (devcouncil / dev)"]
+        CLI["Host CLI\ncmd/devcouncil"]
+        MCP["Host MCP Server\ndevcouncil mcp"]
+        Integ["Integrator\ndevcouncil integrate"]
+        Skills["Skills Engine\ndevcouncil skills"]
+        VerifyCmd["Verify Command\ndevcouncil verify"]
+    end
+
+    subgraph RustAnalysis["Rust Analysis & State Suite"]
+        DevMap["devmap\nCode Graph & 36+ Language ASTs"]
+        DCStore["dcstore\nAtomic Task Leases & SQLite Store"]
+        DCVerify["dcverify\nUnified Diff Parser & Rigor Gates"]
+        DCGrep["dcgrep\nRipgrep Engine & Trigram Index"]
+    end
+
+    Agents <-->|MCP Protocol| MCP
+    Agents <-->|Direct / Tool Calls| CLI
+    Manvi -->|Wraps: imports Go, spawns binaries| Host
+    Manvi -->|Spawns selected modules| RustAnalysis
+    GitPulse -->|Vendors selected crates| DevMap
+    GitPulse -->|manvi serve| Manvi
+
+    CLI -->|Execs| DevMap
+    CLI -->|Integrates| Integ
+    CLI -->|Scaffolds| Skills
+    VerifyCmd -->|Scope, orphan, expected tests| DCStore
+    Manvi -->|Spawns dcverify rigor| DCVerify
+
+    MCP -->|Leases and tasks| DCStore
+    MCP -->|verify_task → Go Run| VerifyCmd
 ```
 
-## Install From Source
+---
 
-For local development inside this checkout:
+## Repository Layout
 
-```bash
-uv sync
-uv run dev --help
 ```
-
-For a global install from this repository:
-
-```bash
-uv tool install --force .
-dev --help
-devcouncil --help
+DevCouncil/
+├── backend/go_orchestrator/     # Go host orchestrator binary and packages
+│   ├── cmd/devcouncil/          # Main entrypoint for devcouncil and dev
+│   ├── devcouncil/              # MCP server, integrate, skills, verify implementations
+│   ├── dc/store/                # Interop client for dcstore binary
+│   └── policy/                  # File-write containment & security policies
+├── rust/                        # Analysis and verification engine crates (dcstore, dcverify, dcgrep, devmap)
+├── bin/                         # Node.js npm shim (bin/devcouncil.js)
+├── scripts/                     # Platform installers (install.sh, install-components.sh)
+└── docs/                        # Architecture, releases, and integration guides
 ```
 
-## Project Shape
-
-DevCouncil implements a 7-phase software-team workflow:
-
-1. Goal analysis and repository mapping.
-2. Requirements drafting.
-3. Council debate and task arbitration.
-4. Gated execution with scoped files and commands.
-5. Deterministic verification.
-6. Repair-loop generation.
-7. Evidence reporting.
-
-Read [docs/architecture.md](docs/architecture.md) for the artifact graph, gating state machine, and component layout.
+---
 
-## Contributions
+## Documentation
 
-Project ideas and execution patterns come from the open-source ecosystem:
+- [Release Notes (v0.2.0)](docs/releases/v0.2.0.md): Native orchestration and analysis migration.
+- [Native cutover follow-ups](docs/TODO.md): Open wiring and honesty work after Phase 7.
+- [Architecture Decisions & Python Retirement](docs/PHASE7_LONG_TAIL.md): Background on the transition to native Go and Rust binaries.
+- [Archived ledgers](docs/archive/README.md): Pre-cutover DevMap plans, audits, and qualification dumps.
+- [Code Graph & DevMap Guide](docs/code-graph.md): Symbol resolution, dead code, and blast radius.
+- [DevMap competitor benchmarks](docs/devmap/README.md#benchmark-comparison): Measured strengths, weaknesses, speed, memory, storage, and correctness against Graphify, Gortex, GitNexus, CodeGraph, codebase-memory-mcp, and ripgrep, with scope limits and raw evidence.
+- [Hero Loop (MCP Closed Loop)](docs/hero-loop.md): Autonomous task loop with deterministic gates.
+- [Coding CLI Integration](docs/coding-cli-integration.md): Configuring Claude, Codex, Cursor, Warp, and Antigravity.
+- [Security Model](docs/security.md): Redaction, write isolation, and containment rules.
+- [Project Status](docs/project-status.md): Subsystem maturity ledger.
 
-- [Sage](https://github.com/usetig/sage): peer-review-first model for planning and critique.
-- [karpathy/llm-council](https://github.com/karpathy/llm-council): for the multi-LLM peer-review pattern.
-- [GPT Pilot](https://github.com/Pythagora-io/gpt-pilot): for role-based software-team concept.
-- [astral-sh/uv](https://github.com/astral-sh/uv): for reproducible Python package/runtime workflows.
-- [OpenHands](https://github.com/All-Hands-AI/OpenHands): for workspace-aware agent execution patterns.
-- [mini-SWE-agent](https://github.com/SWE-agent/mini-swe-agent): for lightweight execution loop inspiration.
-- [SWE-agent](https://github.com/SWE-agent/SWE-agent): for full-spectrum autonomous SWE-style tasking patterns.
-- [GitNexus](https://github.com/abhigyanpatwari/GitNexus): for structural codebase awareness.
-- [graphify](https://github.com/safishamsi/graphify): for knowledge-graph-based coordination concepts.
+---
 
 ## License
 
 Licensed under the **Apache License, Version 2.0**. See [LICENSE](LICENSE) for details.
-
----
-
-**"Trust the model, but verify the graph."**