intake-ai-cli 0.2.0__tar.gz → 0.4.0__tar.gz

{intake_ai_cli-0.2.0 → intake_ai_cli-0.4.0}/CHANGELOG.md

@@ -5,7 +5,142 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-## [0.2.0] - 2026-03-03
+## [0.4.0] - 2026-03-05
+
+### Added
+
+#### MCP (Model Context Protocol) server
+
+- **MCP server module** (`mcp/`): Full MCP server implementation with stdio and SSE transports. Requires `pip install intake-ai-cli[mcp]`.
+- **`create_server()`** (`mcp/server.py`): Creates and configures the MCP server with all tools, resources, and prompts registered. `MCP_SERVER_NAME = "intake-spec"`.
+- **`run_stdio()`** and **`run_sse()`** (`mcp/server.py`): Two transport modes — stdio for CLI agent integration, SSE (HTTP) for browser/IDE integration via starlette + uvicorn.
+- **7 MCP tools** (`mcp/tools.py`):
+  - `intake_show`: Show spec summary with truncated file content (MAX_SECTION_LENGTH = 3000).
+  - `intake_get_context`: Read `context.md` for a spec.
+  - `intake_get_tasks`: List tasks with status filtering (all/pending/in_progress/done/blocked).
+  - `intake_update_task`: Update task status with optional note.
+  - `intake_verify`: Run acceptance checks with optional tag filtering.
+  - `intake_feedback`: Run verification + analyze failures.
+  - `intake_list_specs`: List available specs (filters directories without `requirements.md`).
+- **MCP resources** (`mcp/resources.py`): Dynamic spec file resources via `intake://specs/{name}/{section}` URIs. Supports 6 sections: requirements, tasks, context, acceptance, design, sources. `FILE_MAP` maps section names to actual filenames.
+- **MCP prompts** (`mcp/prompts.py`): Two structured prompt templates:
+  - `implement_next_task`: Reads spec files and generates implementation instructions referencing MCP tools.
+  - `verify_and_fix`: Generates a verify → fix → re-verify loop until all checks pass.
+- **MCPError exception** (`mcp/__init__.py`): Error with `reason` and `suggestion` attributes.
+- **Lazy imports**: `mcp`, `starlette`, `uvicorn` imported lazily with clear `ImportError` messages and installation commands.
+
+#### Watch mode (file monitoring + auto-verification)
+
+- **Watch module** (`watch/`): File watcher with selective re-verification. Requires `pip install intake-ai-cli[watch]`.
+- **SpecWatcher** (`watch/watcher.py`): Monitors project directory using `watchfiles` (Rust-based, efficient). On file change, re-runs verification checks and displays results.
+  - `run_once()`: Single verification without watching.
+  - `run()`: Continuous watch loop with debouncing.
+  - `_filter_ignored()`: Filters files by ignore patterns (fnmatch per path component).
+  - `_matches_any()`: Static method for pattern matching against individual path components.
+  - `_extract_changed_files()`: Extracts relative paths from watchfiles change sets.
+  - `MAX_CHANGED_FILES_DISPLAY = 5`: Limits terminal output.
+- **WatchError exception** (`watch/__init__.py`): Error with `reason` and `suggestion` attributes.
+- **Debouncing**: Configurable via `WatchConfig.debounce_seconds`, passed to watchfiles native debouncing.
+
+#### Configuration
+
+- **MCPConfig** (`config/schema.py`): `specs_dir`, `project_dir`, `transport` (stdio/sse), `sse_port`.
+- **WatchConfig** (`config/schema.py`): `debounce_seconds` (default: 2.0), `ignore_patterns` (default: `["*.pyc", "__pycache__", ".git", "node_modules", ".intake"]`).
+- Both added to `IntakeConfig` as `mcp` and `watch` fields.
+
+#### CLI commands
+
+- **`intake mcp serve`**: Start the MCP server. Options: `--transport` (stdio/sse), `--port`, `--specs-dir`, `--project-dir`.
+- **`intake watch`**: Watch project files and re-run verification. Options: `--project-dir`, `--tags`, `--debounce`, `--verbose`.
+
+#### Optional dependencies
+
+- `mcp = ["mcp[cli]>=1.0"]`: MCP server support.
+- `watch = ["watchfiles>=1.0"]`: Watch mode support.
+- `all = [connectors + watch + mcp]`: Install everything.
+
+#### Test suite
+
+- **772 tests** (up from 673), **0 failures**, **10 skipped** (MCP prompts tests when mcp package not installed). 99 new tests added.
+- 5 new test files: `test_mcp/test_tools.py` (31), `test_mcp/test_resources.py` (17), `test_mcp/test_prompts.py` (10), `test_mcp/test_server.py` (10), `test_watch/test_watcher.py` (27).
+- Config tests: 2 new tests for MCPConfig and WatchConfig nested overrides.
+- CLI tests: 4 new tests for MCP and Watch help output.
+
+### Fixed
+
+- **`_filter_ignored` path matching** (`watch/watcher.py`): Fixed fnmatch to check individual path components (e.g., `.git` now correctly matches `.git/objects/abc`). Added `_matches_any()` static method.
+- **Silent exception in `_handle_get_tasks`** (`mcp/tools.py`): Added `logger.debug("task_state_manager_fallback", ...)` to bare `except Exception` block.
+- **Missing `run_sse` export** (`mcp/__init__.py`): Added `run_sse` to `__all__` and convenience re-export function.
+- **`__all__` sort order** (`mcp/__init__.py`): Fixed RUF022 by sorting alphabetically.
+- **ruff TC003 in tests**: Added `per-file-ignores` rule for `tests/**/*.py` to allow stdlib imports at runtime in test fixtures.
+
+## [0.3.0] - 2026-03-04
+
+### Added
+
+#### 3 API connectors (live data fetching)
+
+- **Jira connector** (`connectors/jira_api.py`): Fetches issues from Jira via REST API. Supports single issue (`jira://PROJ-123`), multiple issues (`jira://PROJ-1,PROJ-2`), JQL queries (`jira://PROJ?jql=sprint=42`), and sprint-based fetching (`jira://PROJ/sprint/42`). Lazy import of `atlassian-python-api`. Saves as JSON temp files compatible with `JiraParser`.
+- **Confluence connector** (`connectors/confluence_api.py`): Fetches pages from Confluence Cloud/Server. Supports page by ID (`confluence://page/123456`), by space and title (`confluence://SPACE/Page-Title`), and CQL search (`confluence://search?cql=...`). Saves as HTML temp files compatible with `ConfluenceParser`.
+- **GitHub connector** (`connectors/github_api.py`): Fetches issues from GitHub repos via PyGithub. Supports single issue (`github://org/repo/issues/42`), multiple issues, and filtered queries (`github://org/repo/issues?labels=bug&state=open`). Max 50 issues, 10 comments per issue.
+- **Doctor connector checks** (`doctor/checks.py`): `_check_connectors()` validates connector credentials when connectors are configured.
+- **3 connector entry points** in `pyproject.toml`: `jira`, `confluence`, `github` under `[project.entry-points."intake.connectors"]`.
+
+#### 4 new exporters (6 total)
+
+- **Claude Code exporter** (`export/claude_code.py`): Generates `CLAUDE.md` (smart append/replace), `.intake/tasks/TASK-NNN.md`, `.intake/verify.sh`, `.intake/spec-summary.md`, and `.intake/spec/` copy.
+- **Cursor exporter** (`export/cursor.py`): Generates `.cursor/rules/intake-spec.mdc` with YAML frontmatter.
+- **Kiro exporter** (`export/kiro.py`): Generates `requirements.md`, `design.md`, `tasks.md` in Kiro's native format.
+- **Copilot exporter** (`export/copilot.py`): Generates `.github/copilot-instructions.md`.
+- All 4 exporters implement the **V2 ExporterPlugin protocol**: `meta`, `supported_agents`, `export() → ExportResult`.
+- **Shared export helpers** (`export/_helpers.py`): `read_spec_file()`, `parse_tasks()`, `load_acceptance_checks()`, `summarize_content()`, `count_requirements()`.
+- **9 new Jinja2 templates**: `claude_md.j2`, `claude_task.md.j2`, `verify_sh.j2`, `cursor_rules.mdc.j2`, `kiro_requirements.md.j2`, `kiro_design.md.j2`, `kiro_tasks.md.j2`, `copilot_instructions.md.j2`, `feedback.md.j2`.
+
+#### Feedback loop (analyze verification failures)
+
+- **Feedback analyzer** (`feedback/analyzer.py`): LLM-based failure analysis with root cause identification, severity classification, and spec amendments. Dataclasses: `SpecAmendment`, `FailureAnalysis`, `FeedbackResult`.
+- **Feedback prompts** (`feedback/prompts.py`): Analysis prompt with `{language}` placeholder and structured JSON output schema.
+- **Suggestion formatter** (`feedback/suggestions.py`): Terminal output (Rich) + agent-specific formatting (generic, claude-code, cursor).
+- **Spec updater** (`feedback/spec_updater.py`): Preview and apply spec amendments with add/modify/remove actions.
+- **`intake feedback` CLI command**: Options `--verify-report`, `--project-dir`, `--apply`, `--agent-format`, `--verbose`.
+
+#### Configuration updates
+
+- **`FeedbackConfig`** model: `auto_amend_spec`, `max_suggestions`, `include_code_snippets`.
+- **Expanded connector configs**: `JiraConfig` (auth_type, fields, max_comments), `ConfluenceConfig` (include_child_pages, max_depth), `GithubConfig` (default_repo).
+- **`ExportConfig`** expanded: `claude_code_task_dir`, `cursor_rules_dir`.
+
+#### Enterprise documentation
+
+- **`docs/seguridad.md`** (NEW): Threat model, data flow, secrets management, redaction patterns, offline/air-gapped mode, audit trail, compliance (SOC2/HIPAA/ISO 27001/GDPR).
+- **`docs/despliegue.md`** (NEW): Docker multi-stage, docker-compose, pre-commit hooks, deployment patterns, env vars.
+- **`docs/integracion-cicd.md`** (NEW): GitHub Actions, GitLab CI, Jenkins, Azure DevOps, JUnit/JSON reports, spec drift, notifications.
+- **`docs/flujos-trabajo.md`** (NEW): Solo developer, small team, enterprise, monorepo, AI agent, regulated industries.
+- **`docs/conectores.md`** (NEW): Jira, Confluence, GitHub API connectors documentation.
+- **`docs/feedback.md`** (NEW): Feedback loop usage, severity levels, spec amendments, agent formats.
+- **`SECURITY.md`** (NEW): English security policy at project root (GitHub Security tab).
+- 10 existing documentation files updated with Phase 2 content.
+
+#### Test suite
+
+- **673 tests** (up from 492), **0 failures**. 181 new tests added.
+- 14 new test files covering connectors, exporters, feedback, and protocol conformance.
+- Protocol conformance tests: 20 parametrized tests for V2 exporters + 3 for connectors.
+
+### Fixed
+
+#### QA Audit (14 mypy errors, 7 ruff errors, 25 format issues)
+
+- **ExporterRegistry type safety** (`export/registry.py`): Introduced `AnyExporter = Exporter | Any` union type for V1/V2 dual protocol support. Changed `_exporters`, `register()`, and `get()` types to accept both V1 and V2 exporters without `type: ignore`.
+- **Connector `Returning Any`** (`connectors/jira_api.py`, `connectors/confluence_api.py`): Wrapped `data.get()` calls with `list()` and explicit `str` annotations to satisfy mypy --strict.
+- **Unused `type: ignore` comments**: Removed 3 stale `# type: ignore[import-untyped]` from connectors after `atlassian-python-api` and `PyGithub` gained type stubs.
+- **CLI report type** (`cli.py`): Changed `report_data: dict[str, object]` to `dict[str, Any]` for correct iterable access.
+- **Connector registration guard** (`cli.py`): Added `isinstance(connector_obj, ConnectorPlugin)` check for type-safe connector registration.
+- **Unused imports** (`tests/`): Removed `PropertyMock`, `pytest`, `json`, `yaml` unused imports in test files.
+- **TYPE_CHECKING import** (`tests/test_feedback/test_analyzer.py`): Moved `Path` import to `TYPE_CHECKING` block (TC003).
+- **25 files reformatted** with `ruff format` for consistent code style.
+
+## [0.2.0] - 2026-03-04
 
 ### Added
 

{intake_ai_cli-0.2.0 → intake_ai_cli-0.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: intake-ai-cli
-Version: 0.2.0
+Version: 0.4.0
 Summary: From requirements in any format to verified implementation
 Project-URL: Homepage, https://diego303.github.io/intake-docs/
 Project-URL: Repository, https://github.com/Diego303/intake-cli
@@ -28,6 +28,11 @@ Requires-Dist: python-docx>=1.1
 Requires-Dist: pyyaml>=6.0
 Requires-Dist: rich>=13.0
 Requires-Dist: structlog>=24.0
+Provides-Extra: all
+Requires-Dist: atlassian-python-api>=3.40; extra == 'all'
+Requires-Dist: mcp>=1.0; extra == 'all'
+Requires-Dist: pygithub>=2.0; extra == 'all'
+Requires-Dist: watchfiles>=0.21; extra == 'all'
 Provides-Extra: connectors
 Requires-Dist: atlassian-python-api>=3.40; extra == 'connectors'
 Requires-Dist: pygithub>=2.0; extra == 'connectors'
@@ -75,7 +80,7 @@ intake processes requirements through a 5-phase pipeline:
 2. **Analyze** — LLM extracts structured requirements, detects conflicts, deduplicates
 3. **Generate** — Produce 6 spec files + `spec.lock.yaml`
 4. **Verify** — Run executable acceptance checks against the implementation
-5. **Export** — Generate agent-ready output (architect, Claude Code, Cursor, generic)
+5. **Export** — Generate agent-ready output (Claude Code, Cursor, Kiro, Copilot, architect, generic)
 
 ### The 6 Spec Files
 
@@ -94,6 +99,12 @@ intake processes requirements through a 5-phase pipeline:
 
 ```bash
 pip install intake-ai-cli
+
+# With optional features
+pip install intake-ai-cli[mcp]         # MCP server support
+pip install intake-ai-cli[watch]       # Watch mode support
+pip install intake-ai-cli[connectors]  # Jira, Confluence, GitHub API connectors
+pip install intake-ai-cli[all]         # Everything
 ```
 
 Requires Python 3.12+. The CLI command is `intake`.
@@ -132,6 +143,29 @@ intake init "Fix login bug" -s notes.txt --mode quick
 # Fetch requirements from a URL
 intake init "API review" -s https://wiki.company.com/rfc/auth
 
+# Fetch from live APIs (requires credentials)
+intake init "Sprint planning" -s jira://PROJ/sprint/42
+intake init "Wiki review" -s confluence://SPACE/Page-Title
+intake init "Bug triage" -s github://org/repo/issues?labels=bug
+
+# Start MCP server for AI agent integration
+intake mcp serve --transport stdio
+intake mcp serve --transport sse --port 8080
+
+# Watch project files and auto-verify on changes
+intake watch ./specs/auth-oauth2 --project-dir . --verbose
+intake watch ./specs/auth-oauth2 --tags test,lint --debounce 3
+
+# Export for specific agents
+intake init "Payments" -s reqs.pdf --format claude-code
+intake export ./specs/auth -f cursor -o .
+intake export ./specs/auth -f kiro -o .
+intake export ./specs/auth -f copilot -o .
+
+# Analyze verification failures and get fix suggestions
+intake feedback ./specs/auth-oauth2
+intake feedback ./specs/auth -r report.json --apply --agent-format claude-code
+
 # List discovered plugins
 intake plugins list
 
@@ -157,9 +191,44 @@ intake task update ./specs/auth-oauth2 1 done --note "Implemented and tested"
 | URLs | `http://`, `https://` | Fetches page, converts HTML → Markdown |
 | Slack export | `.json` (auto-detected) | Messages, threads, decisions, action items |
 | GitHub Issues | `.json` (auto-detected) | Issues, labels, comments, cross-references |
+| **Jira API** | `jira://PROJ-123` | Live issue fetching via REST API |
+| **Confluence API** | `confluence://SPACE/Title` | Live page fetching via REST API |
+| **GitHub API** | `github://org/repo/issues/42` | Live issue fetching via PyGithub |
 
 Format is auto-detected by file extension and content inspection. Jira, Slack, and GitHub Issues JSON exports are distinguished automatically from generic JSON files. Confluence HTML is distinguished from generic HTML.
 
+### Live API Connectors
+
+Connect directly to project management tools (requires credentials):
+
+```bash
+# Jira: single issue, multiple, JQL, sprint
+intake init "Sprint" -s jira://PROJ-123
+intake init "Sprint" -s "jira://PROJ?jql=sprint=42"
+
+# Confluence: page by ID, by space/title, CQL search
+intake init "Docs" -s confluence://page/123456
+intake init "Docs" -s confluence://SPACE/Page-Title
+
+# GitHub: single/multiple issues, filtered queries
+intake init "Bugs" -s github://org/repo/issues/42
+intake init "Bugs" -s "github://org/repo/issues?labels=bug&state=open"
+```
+
+Configure credentials in `.intake.yaml`:
+
+```yaml
+connectors:
+  jira:
+    url: https://your-org.atlassian.net
+    # Set JIRA_API_TOKEN and JIRA_EMAIL env vars
+  confluence:
+    url: https://your-org.atlassian.net/wiki
+    # Set CONFLUENCE_API_TOKEN and CONFLUENCE_EMAIL env vars
+  github:
+    # Set GITHUB_TOKEN env var
+```
+
 ---
 
 ## Commands
@@ -175,10 +244,14 @@ Format is auto-detected by file extension and content inspection. Jira, Slack, a
 | `intake diff` | Compare two spec versions | **Available** |
 | `intake doctor` | Check environment and configuration health | **Available** |
 | `intake doctor --fix` | Auto-fix environment issues (install deps, create config) | **Available** |
-| `intake plugins list` | List all discovered plugins (parsers, exporters) | **Available** |
+| `intake feedback` | Analyze verification failures and suggest fixes | **Available** |
+| `intake feedback --apply` | Auto-apply suggested spec amendments | **Available** |
+| `intake plugins list` | List all discovered plugins (parsers, exporters, connectors) | **Available** |
 | `intake plugins check` | Validate plugin compatibility | **Available** |
 | `intake task list` | List tasks from a spec with current status | **Available** |
 | `intake task update` | Update a task's status (pending/in_progress/done/blocked) | **Available** |
+| `intake mcp serve` | Start MCP server (stdio or SSE transport) | **Available** |
+| `intake watch` | Watch project files and re-run verification on changes | **Available** |
 
 ---
 
@@ -205,7 +278,34 @@ spec:
   auto_mode: true              # auto-detect quick/standard/enterprise
 
 export:
-  default_format: generic      # architect | claude-code | cursor | kiro | generic
+  default_format: generic      # architect | claude-code | cursor | kiro | copilot | generic
+
+feedback:
+  auto_amend_spec: false       # Auto-apply spec amendments from feedback
+  max_suggestions: 10          # Max suggestions per analysis
+  include_code_snippets: true  # Include code examples in suggestions
+
+connectors:
+  jira:
+    url: https://your-org.atlassian.net
+  confluence:
+    url: https://your-org.atlassian.net/wiki
+  github: {}                   # Uses GITHUB_TOKEN env var
+
+mcp:
+  specs_dir: ./specs           # Where specs live
+  project_dir: .               # Project root for verification
+  transport: stdio             # stdio | sse
+  sse_port: 8080               # Port for SSE transport
+
+watch:
+  debounce_seconds: 2.0        # Wait before re-running verification
+  ignore_patterns:             # Files/dirs to ignore
+    - "*.pyc"
+    - "__pycache__"
+    - ".git"
+    - "node_modules"
+    - ".intake"
 ```
 
 ### Presets
@@ -245,7 +345,7 @@ See the [`examples/`](examples/) directory for ready-to-run scenarios:
 src/intake/
 ├── cli.py                  # Click CLI — thin adapter, no logic
 ├── config/                 # Pydantic v2 models, presets, layered loader
-│   ├── schema.py           #   7 config models (LLM, Project, Spec, Verification, Export, Security, Connectors)
+│   ├── schema.py           #   11 config models (LLM, Project, Spec, Verification, Export, Security, Connectors, Feedback, MCP, Watch)
 │   ├── presets.py           #   minimal / standard / enterprise presets
 │   ├── loader.py            #   Layered merge: defaults → preset → YAML → CLI
 │   └── defaults.py          #   Centralized constants
@@ -253,8 +353,11 @@ src/intake/
 │   ├── protocols.py         #   V2 protocols: ParserPlugin, ExporterPlugin, ConnectorPlugin
 │   ├── discovery.py         #   Entry point scanning via importlib.metadata
 │   └── hooks.py             #   Pipeline hook system (HookManager)
-├── connectors/             # Connector infrastructure (Phase 2 prep)
-│   └── base.py              #   ConnectorRegistry, ConnectorError
+├── connectors/             # Live API connectors
+│   ├── base.py              #   ConnectorRegistry, ConnectorError
+│   ├── jira_api.py          #   Jira REST API (single/multi/JQL/sprint)
+│   ├── confluence_api.py    #   Confluence REST API (page/space/CQL)
+│   └── github_api.py        #   GitHub API via PyGithub (issues/filters)
 ├── ingest/                 # Phase 1 — 11 parsers, registry, auto-detection
 │   ├── base.py              #   ParsedContent dataclass + Parser Protocol
 │   ├── registry.py          #   Auto-detection + plugin discovery + parser dispatch
@@ -287,24 +390,50 @@ src/intake/
 ├── verify/                 # Phase 4 — Acceptance check engine
 │   ├── engine.py           #   4 check types: command, files_exist, pattern_*
 │   └── reporter.py         #   Terminal (Rich), JSON, JUnit XML reporters
-├── export/                 # Phase 5 — Agent-ready output
+├── export/                 # Phase 5 — Agent-ready output (6 exporters)
 │   ├── base.py             #   Exporter Protocol
 │   ├── registry.py         #   Plugin discovery + format-based dispatch
+│   ├── _helpers.py         #   Shared utilities (parse_tasks, load_checks, etc.)
+│   ├── claude_code.py      #   CLAUDE.md + tasks + verify.sh
+│   ├── cursor.py           #   .cursor/rules/intake-spec.mdc
+│   ├── kiro.py             #   Kiro-native requirements/design/tasks
+│   ├── copilot.py          #   .github/copilot-instructions.md
 │   ├── architect.py        #   pipeline.yaml generation
 │   └── generic.py          #   SPEC.md + verify.sh generation
 ├── diff/                   # Spec comparison
 │   └── differ.py           #   Compare two specs by requirement/task IDs
+├── mcp/                    # MCP server (Model Context Protocol)
+│   ├── __init__.py         #   MCPError + re-exports
+│   ├── server.py           #   Server creation + stdio/SSE transports
+│   ├── tools.py            #   7 tools: show, context, tasks, update, verify, feedback, list
+│   ├── resources.py        #   Dynamic spec file resources (intake://specs/{name}/{section})
+│   └── prompts.py          #   Prompt templates: implement_next_task, verify_and_fix
+├── watch/                  # Watch mode (file monitoring + auto-verification)
+│   ├── __init__.py         #   WatchError exception
+│   └── watcher.py          #   SpecWatcher with watchfiles integration
+├── feedback/               # Feedback loop (analyze failures, suggest fixes)
+│   ├── analyzer.py         #   LLM-based failure analysis
+│   ├── prompts.py          #   Feedback analysis prompt
+│   ├── suggestions.py     #   Multi-format suggestion formatter
+│   └── spec_updater.py    #   Preview + apply spec amendments
 ├── doctor/                 # Environment health checks
-│   └── checks.py            #   Python, API keys, deps, config validation
+│   └── checks.py            #   Python, API keys, deps, connectors, config validation
 ├── llm/                    # LiteLLM wrapper (used by analyze/ only)
 │   └── adapter.py           #   Async completion, retry, cost tracking, budget
-├── templates/              # Jinja2 templates for spec generation
+├── templates/              # Jinja2 templates (15 total)
 │   ├── requirements.md.j2   #   FR, NFR, conflicts, open questions
 │   ├── design.md.j2         #   Components, files, tech decisions
 │   ├── tasks.md.j2          #   Task summary + status + detailed sections
 │   ├── acceptance.yaml.j2   #   Executable acceptance checks
 │   ├── context.md.j2        #   Project context for agents
-│   └── sources.md.j2        #   Source traceability mapping
+│   ├── sources.md.j2        #   Source traceability mapping
+│   ├── claude_md.j2         #   Claude Code CLAUDE.md spec section
+│   ├── claude_task.md.j2    #   Claude Code per-task file
+│   ├── verify_sh.j2         #   Claude Code verification script
+│   ├── cursor_rules.mdc.j2  #   Cursor rules file
+│   ├── kiro_*.md.j2         #   Kiro requirements/design/tasks (3 files)
+│   ├── copilot_instructions.md.j2  # Copilot instructions
+│   └── feedback.md.j2      #   Feedback results template
 └── utils/                  # Shared utilities
     ├── file_detect.py       #   Extension-based format detection
     ├── project_detect.py    #   Auto-detect tech stack from project files
@@ -317,7 +446,7 @@ src/intake/
 **Key design principles:**
 
 - **Protocol over ABC** — All extension points use `typing.Protocol`
-- **Plugin-first architecture** — Parsers and exporters discovered via entry points, manual fallback
+- **Plugin-first architecture** — Parsers, exporters, and connectors discovered via entry points, manual fallback
 - **Dataclasses for pipeline data, Pydantic for config** — Never mixed
 - **Async only in analyze/** — Everything else is synchronous
 - **Offline mode** — Parsing, verification, export, diff, doctor all work without LLM
@@ -340,7 +469,84 @@ architect pipeline specs/auth-system/pipeline.yaml
 
 ```bash
 intake init "Payments" -s reqs.pdf --format claude-code
-# Generates CLAUDE.md + tasks + verify.sh
+# Generates CLAUDE.md + .intake/tasks/ + .intake/verify.sh + .intake/spec-summary.md
+```
+
+### With Cursor
+
+```bash
+intake export ./specs/auth -f cursor -o .
+# Generates .cursor/rules/intake-spec.mdc (auto-loaded by Cursor)
+```
+
+### With Kiro
+
+```bash
+intake export ./specs/auth -f kiro -o .
+# Generates requirements.md, design.md, tasks.md in Kiro native format
+```
+
+### With GitHub Copilot
+
+```bash
+intake export ./specs/auth -f copilot -o .
+# Generates .github/copilot-instructions.md (auto-loaded by Copilot)
+```
+
+### MCP Server (AI Agent Integration)
+
+intake exposes specs via the [Model Context Protocol](https://modelcontextprotocol.io/), allowing AI agents to consume specs in real time:
+
+```bash
+# Start MCP server with stdio transport (for CLI agents like Claude Code)
+intake mcp serve --transport stdio
+
+# Start MCP server with SSE transport (for IDE integrations)
+intake mcp serve --transport sse --port 8080
+```
+
+**MCP tools available to agents:**
+- `intake_show` — View spec summary
+- `intake_get_context` — Read project context
+- `intake_get_tasks` — List tasks with status filtering
+- `intake_update_task` — Mark tasks as done/in_progress
+- `intake_verify` — Run acceptance checks
+- `intake_feedback` — Analyze verification failures
+- `intake_list_specs` — List available specs
+
+**MCP resources:** Direct access to spec files via `intake://specs/{name}/{section}` URIs.
+
+**MCP prompts:** `implement_next_task` and `verify_and_fix` provide structured starting points for agents.
+
+Install MCP support: `pip install intake-ai-cli[mcp]`
+
+### Watch Mode
+
+Automatically re-run verification checks when project files change:
+
+```bash
+# Watch with default settings
+intake watch ./specs/auth-oauth2 --project-dir .
+
+# Watch with tags filter and custom debounce
+intake watch ./specs/auth-oauth2 --tags test,lint --debounce 3 --verbose
+```
+
+Uses `watchfiles` (Rust-based) for efficient file monitoring with configurable debounce and ignore patterns.
+
+Install watch support: `pip install intake-ai-cli[watch]`
+
+### Feedback Loop
+
+```bash
+# Analyze why verification checks failed and get fix suggestions
+intake feedback ./specs/auth-oauth2
+
+# Use a previous report and auto-apply spec amendments
+intake feedback ./specs/auth -r report.json --apply
+
+# Get suggestions formatted for your agent
+intake feedback ./specs/auth --agent-format claude-code
 ```
 
 ### With CI/CD
@@ -374,7 +580,7 @@ ruff format src/ tests/
 mypy src/ --strict
 ```
 
-Current test suite: **492 tests**, **86% coverage**, **0 mypy --strict errors**, **0 ruff warnings**.
+Current test suite: **772 tests**, **0 mypy --strict errors**, **0 ruff warnings**.
 
 ### Implementation Status
 
@@ -384,12 +590,15 @@ Current test suite: **492 tests**, **86% coverage**, **0 mypy --strict errors**,
 | Phase 2 — Analyze | `analyze/` (orchestrator + 7 sub-modules + complexity) | Implemented |
 | Phase 3 — Generate | `generate/` (spec builder + adaptive builder + 6 templates + lock) | Implemented |
 | Phase 4 — Verify | `verify/` (engine + 3 reporters) | Implemented |
-| Phase 5 — Export | `export/` (architect + generic + plugin registry) | Implemented |
+| Phase 5 — Export | `export/` (6 exporters: claude-code, cursor, kiro, copilot, architect, generic) | Implemented |
 | Plugins | `plugins/` (protocols + discovery + hooks) | Implemented |
-| Connectors | `connectors/` (registry infrastructure, no concrete connectors) | Implemented |
+| Connectors | `connectors/` (Jira, Confluence, GitHub API connectors) | Implemented |
+| Feedback | `feedback/` (analyzer + suggestions + spec updater) | Implemented |
+| MCP Server | `mcp/` (server + 7 tools + resources + prompts) | Implemented |
+| Watch Mode | `watch/` (SpecWatcher with watchfiles) | Implemented |
 | Standalone | `doctor/`, `config/`, `llm/`, `utils/` | Implemented |
 | Standalone | `diff/` (spec differ) | Implemented |
-| CLI | 13 commands/subcommands wired end-to-end | Implemented |
+| CLI | 19 commands/subcommands wired end-to-end | Implemented |
 
 ---