diff --git a/.claude/skills/gaia-release/SKILL.md b/.claude/skills/gaia-release/SKILL.md
index f161fe3f1..fb11b8b54 100644
--- a/.claude/skills/gaia-release/SKILL.md
+++ b/.claude/skills/gaia-release/SKILL.md
@@ -53,7 +53,7 @@ These map to [CLAUDE.md](CLAUDE.md). Re-read them whenever this skill runs.
 
 - **No Claude attribution anywhere** — not in PR titles, PR bodies, commit messages (no `Co-Authored-By: Claude ...` trailer), release notes, code comments, or the Discord announcement.
 - **No silent fallbacks** — if a validator fails, a step times out, or a workflow run isn't found, stop with an actionable error. Do not retry blindly. Do not "proceed anyway."
-- **Match house style for release notes** — see *Generation parameters* in Phase 1. In short: value-prop first, **local agents are the headline** (not the SDK), one agent/command per highlight, plain language, engaging but factual, and **no emoji, no fluff** ("finally", "silently", "no more crashes", "we're excited to announce", "blazing", "Here's the good stuff"). Read the **last 2–3 release notes** before drafting. Patch releases do **not** include a `pip install` block. Use the `Why upgrade:` framing with a short bullet list, then `## What's New`, then `## Bug Fixes`, then `## Full Changelog`.
+- **Match house style for release notes** — see *Generation parameters* in Phase 1. In short: value-prop first, **local agents are the headline** (not the SDK), one agent/command per highlight, plain language, engaging but factual, and **no emoji, no fluff** (full banned-phrase list under *Generation parameters*). Read the **last 2–3 release notes** before drafting. Patch releases do **not** include a `pip install` block. Use the `Why upgrade:` framing with a short bullet list, then `## What's New`, then `## Bug Fixes`, then `## Full Changelog`.
 - **Match the previous release PR body shape exactly** — read the most recent merged `Release vX.Y.Z` PR (e.g. `gh pr list --repo amd/gaia --state merged --search "Release v in:title" --limit 3`). Open with `# GAIA vX.Y.Z Release Notes` (no MDX frontmatter in the PR body), end with a `Release checklist` section. Style drift here costs review cycles.
 - **Bulletproof commits only** — every change made by this skill must satisfy the four criteria in CLAUDE.md (validated, critiqued, scope-clean, no half-finished work) before being committed.
 - **Pushing tags is irreversible.** Always confirm the SHA the tag will point to and the green status of the pre-tag verification run before `git push origin v<version>`.
@@ -90,12 +90,12 @@ These map to [CLAUDE.md](CLAUDE.md). Re-read them whenever this skill runs.
 
    If the requested version doesn't match the rubric, **stop and surface the mismatch**: *"You asked for `v<requested>` (patch). I see N feat commits since `<prev>` including `<one or two examples>` — this looks minor-shaped. Continue as patch, or bump to `v<suggested>`?"* Do not silently proceed.
 
-2. **Read the last 2–3 release notes** to match style and length.
+2. **Read the last 2–3 release notes** to match structure and length (not tone — see *Generation parameters*).
    - [docs/releases/v0.17.4.mdx](docs/releases/v0.17.4.mdx)
    - [docs/releases/v0.17.3.mdx](docs/releases/v0.17.3.mdx)
    - [docs/releases/v0.17.2.mdx](docs/releases/v0.17.2.mdx)
 
-   Cross-check: same frontmatter shape, same section headings, same tone, same level of detail per entry. Patch releases are short; minor/major releases include `pip install` and may have a "Highlights" block.
+   Cross-check: same frontmatter shape, same section headings, same *structure* and length per entry — but **not** the prior tone. The last few releases predate the *Generation parameters* below; match their shape, not their dryness. Patch releases are short; minor/major releases include `pip install` and may have a "Highlights" block.
 
 3. **Create [docs/releases/v<version>.mdx](docs/releases/)** with this skeleton (adapt to whether it's patch / minor / major):
 
@@ -121,7 +121,9 @@ These map to [CLAUDE.md](CLAUDE.md). Re-read them whenever this skill runs.
 
    <Lead with the outcome and why it matters, in plain language that makes the reader
    want to try it. Then one line on how to run it, PR linked inline. One agent or
-   command per entry — add another `### ` block for the next one.>
+   command per entry — add another `### ` block for the next one. Not every highlight
+   is a command — for UI / SDK / perf items, use a plain title with no trailing
+   command.>
 
    ---
 
@@ -148,7 +150,7 @@ These map to [CLAUDE.md](CLAUDE.md). Re-read them whenever this skill runs.
 
    - **Value-prop first.** Open each entry with what the user can now do and why it
      matters — the outcome, not the implementation. "Triage your inbox in one command"
-     before "added EmailAgent with IMAP polling".
+     before "added EmailAgent with Gmail polling".
    - **Local agents are the headline.** Lead with the agents that solve real problems
      (`gaia browse`, `gaia analyze`, email triage, …); SDK / infra / refactors are
      supporting detail. People come for the agents, not the SDK.
@@ -167,13 +169,13 @@ These map to [CLAUDE.md](CLAUDE.md). Re-read them whenever this skill runs.
 
    > **Bad** (dry, implementation-first, no reason to care):
    > ### EmailAgent
-   > Adds an EmailAgent with IMAP polling and a rules engine for classification.
+   > Adds an EmailAgent with Gmail polling and a rules engine for classification.
 
    > **Good** (value-first, plain, makes you want to try it):
    > ### Triage your inbox from the terminal — `gaia email`
    > Point GAIA at your inbox and it sorts the noise from what needs you: drafts
    > replies to routine mail, flags what's urgent, leaves the rest. Runs locally, so
-   > your mail never leaves your machine. Try it: `gaia email triage`.
+   > your mail never leaves your machine. Try it: `gaia email`.
 
 4. **Update [docs/docs.json](docs/docs.json):**
    - Add `releases/v<version>` to the Releases tab.
diff --git a/CLAUDE.md b/CLAUDE.md
index a1cfc9d5c..d97e60413 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -307,7 +307,7 @@ gaia eval agent --category rag_quality --agent-type doc \
 ps aux | grep "gaia eval" | grep -v grep | wc -l    # must print "0"
 ```
 
-This applies to **all** eval flavours: `gaia eval agent`, `gaia eval --use-claude`, `gaia eval fix-code`, batch experiments. The judge LLM (Claude) can run concurrently across scenarios — the bottleneck is the local Lemonade backend, which is single-tenant per model slot.
+This applies to every `gaia eval agent` run — including `--fix` auto-fix runs and any batch fix-loop that chains them. The judge LLM (Claude) can run concurrently across scenarios — the bottleneck is the local Lemonade backend, which is single-tenant per model slot.
 
 ## Development Workflow
 
@@ -367,13 +367,20 @@ gaia/
 │   │   ├── chat/       # ChatAgent with RAG (tools/rag_tools, tools/shell_tools)
 │   │   ├── code/       # CodeAgent with orchestration, validators, file_io tools
 │   │   ├── builder/    # BuilderAgent — scaffolds new agents from templates
-│   │   ├── summarize/  # SummarizeAgent — document/text summarization
+│   │   ├── summarize/  # SummarizerAgent — document/text summarization
 │   │   ├── blender/    # BlenderAgent for 3D automation
 │   │   ├── jira/       # JiraAgent for issue management
 │   │   ├── docker/     # DockerAgent for containerization
 │   │   ├── emr/        # MedicalIntakeAgent for healthcare (VLM)
 │   │   ├── routing/    # RoutingAgent for intelligent agent selection
 │   │   ├── sd/         # SDAgent for Stable Diffusion image generation
+│   │   ├── email/      # EmailTriageAgent for Gmail (local inference)
+│   │   ├── browser/    # BrowserAgent — web research (gaia browse)
+│   │   ├── analyst/    # AnalystAgent — structured data analysis (gaia analyze)
+│   │   ├── docqa/      # DocumentQAAgent — RAG document Q&A
+│   │   ├── fileio/     # FileIOAgent — file system + safe shell
+│   │   ├── code_index/ # Code-aware indexing/search agent + mixin
+│   │   ├── connectors_demo/ # ConnectorsDemoAgent — Google/GitHub connector demo
 │   │   └── registry.py # Agent registry + KNOWN_TOOLS map
 │   ├── api/            # OpenAI-compatible REST API server
 │   ├── apps/           # Standalone applications
@@ -386,21 +393,28 @@ gaia/
 │   │   └── _shared/    # Shared assets for apps
 │   ├── audio/          # Audio processing (Whisper ASR, Kokoro TTS)
 │   ├── chat/           # Agent SDK (AgentSDK class, prompts, app entry)
+│   ├── code_index/     # Code indexing/search backend
+│   ├── connectors/     # Connector framework (Google/GitHub OAuth, MCP-server connectors, grants)
 │   ├── database/       # DatabaseMixin and DatabaseAgent
 │   ├── electron/       # Electron app integration
 │   ├── eval/           # Evaluation framework
+│   ├── filesystem/     # Filesystem service/utilities
+│   ├── governance/     # Governance / guardrails layer
 │   ├── img/            # Shared image assets
 │   ├── installer/      # Install/init commands (gaia init, lemonade installer)
 │   ├── llm/            # LLM backend clients (Lemonade, Claude, OpenAI) + providers/
 │   ├── mcp/            # Model Context Protocol servers/clients
+│   ├── messaging/      # Messaging adapters (Telegram, …)
 │   ├── rag/            # Document retrieval (RAG)
 │   ├── sd/             # Stable Diffusion tool mixin (SDToolsMixin)
+│   ├── scratchpad/     # Scratchpad tables backend
 │   ├── shell/          # Shell integration
 │   ├── talk/           # Voice interaction SDK
 │   ├── testing/        # Test utilities and fixtures
 │   ├── ui/             # Agent UI backend (FastAPI server, routers, SSE, database)
 │   ├── utils/          # Utility modules (FileWatcher, parsing)
 │   ├── vlm/            # Vision LLM tool mixin (VLMToolsMixin, structured extraction)
+│   ├── web/            # Web utilities (search/fetch backend)
 │   └── cli.py          # Main CLI entry point (all `gaia <command>` subparsers)
 ├── tests/              # Test suite
 │   ├── unit/           # Unit tests
@@ -454,16 +468,21 @@ Defined in [`setup.py`](setup.py) under `console_scripts`:
 
 | Agent | Location | Description | Default Model |
 |-------|----------|-------------|---------------|
-| **ChatAgent** | `agents/chat/agent.py` | Document Q&A with RAG | Qwen3.5-35B |
-| **CodeAgent** | `agents/code/agent.py` | Code generation with orchestration | Qwen3.5-35B |
-| **BuilderAgent** | `agents/builder/agent.py` | Scaffolds new agents from templates | Qwen3.5-35B |
-| **SummarizeAgent** | `agents/summarize/agent.py` | Document/text summarization | Qwen3.5-35B |
-| **JiraAgent** | `agents/jira/agent.py` | Jira issue management | Qwen3.5-35B |
-| **BlenderAgent** | `agents/blender/agent.py` | 3D scene automation | Qwen3.5-35B |
-| **DockerAgent** | `agents/docker/agent.py` | Container management | Qwen3.5-35B |
+| **ChatAgent** | `agents/chat/agent.py` | Document Q&A with RAG | Gemma-4-E4B |
+| **CodeAgent** | `agents/code/agent.py` | Code generation with orchestration | Qwen3.5-35B-A3B |
+| **BuilderAgent** | `agents/builder/agent.py` | Scaffolds new agents from templates | Qwen3.5-35B-A3B |
+| **SummarizerAgent** | `agents/summarize/agent.py` | Document/text summarization | Gemma-4-E4B |
+| **JiraAgent** | `agents/jira/agent.py` | Jira issue management | Qwen3.5-35B-A3B |
+| **BlenderAgent** | `agents/blender/agent.py` | 3D scene automation | Gemma-4-E4B |
+| **DockerAgent** | `agents/docker/agent.py` | Container management | Gemma-4-E4B |
 | **MedicalIntakeAgent** | `agents/emr/agent.py` | Medical form processing | Qwen3-VL-4B (VLM) |
-| **RoutingAgent** | `agents/routing/agent.py` | Intelligent agent selection | Qwen3.5-35B |
+| **RoutingAgent** | `agents/routing/agent.py` | Intelligent agent selection | Qwen3.5-35B-A3B (`AGENT_ROUTING_MODEL`) |
 | **SDAgent** | `agents/sd/agent.py` | Stable Diffusion image generation | SDXL-Turbo |
+| **EmailTriageAgent** | `agents/email/agent.py` | Email triage for Gmail — local inference, needs the Google connector | Gemma-4-E4B |
+| **BrowserAgent** | `agents/browser/agent.py` | Web research — search, page fetch, download (`gaia browse`) | Gemma-4-E4B |
+| **AnalystAgent** | `agents/analyst/agent.py` | Structured data analysis with scratchpad tables (`gaia analyze`) | Gemma-4-E4B |
+
+`gaia browse` and `gaia analyze` invoke BrowserAgent and AnalystAgent respectively (see [`src/gaia/cli.py`](src/gaia/cli.py)). `gaia telegram` is a messaging adapter, not an agent. Internal building-block agents (DocumentQAAgent, FileIOAgent, ConnectorsDemoAgent) live under `src/gaia/agents/` but aren't standalone CLI commands.
 
 ### Agent Registry & Tool Mixins
 
@@ -472,19 +491,24 @@ New agents are Python classes inheriting from `Agent` (see [`src/gaia/agents/bas
 | Tool name | Mixin | Purpose |
 |-----------|-------|---------|
 | `rag` | `gaia.agents.chat.tools.rag_tools.RAGToolsMixin` | Document retrieval |
+| `code_index` | `gaia.agents.code_index.tools.mixin.CodeIndexToolsMixin` | Code-aware indexing/search |
 | `file_search` | `gaia.agents.tools.file_tools.FileSearchToolsMixin` | Fuzzy/glob file search |
 | `file_io` | `gaia.agents.code.tools.file_io.FileIOToolsMixin` | Read/write/edit files |
 | `shell` | `gaia.agents.chat.tools.shell_tools.ShellToolsMixin` | Sandboxed shell commands |
 | `screenshot` | `gaia.agents.tools.screenshot_tools.ScreenshotToolsMixin` | Screen capture |
+| `filesystem` | `gaia.agents.tools.filesystem_tools.FileSystemToolsMixin` | Filesystem operations |
+| `scratchpad` | `gaia.agents.tools.scratchpad_tools.ScratchpadToolsMixin` | Scratchpad tables/notes |
+| `browser` | `gaia.agents.tools.browser_tools.BrowserToolsMixin` | Web search / page fetch / download |
 | `sd` | `gaia.sd.mixin.SDToolsMixin` | Stable Diffusion image generation |
 | `vlm` | `gaia.vlm.mixin.VLMToolsMixin` | Vision LLM / structured extraction |
 
 When adding a new tool mixin, register it in `KNOWN_TOOLS` so other agents can compose it by name.
 
 ### Default Models
-- General tasks: `Qwen3-0.6B-GGUF`
-- Code/Agents: `Qwen3.5-35B-A3B-GGUF`
+- Default for most agents and `gaia llm`: `Gemma-4-E4B-it-GGUF` (`DEFAULT_MODEL_NAME` in [`src/gaia/llm/lemonade_client.py`](src/gaia/llm/lemonade_client.py))
+- Code-heavy agents (Code, Builder, Jira): `Qwen3.5-35B-A3B-GGUF` (hardcoded per agent)
 - Vision tasks: `Qwen3-VL-4B-Instruct-GGUF`
+- Image generation (SD): `SDXL-Turbo`
 
 ## CLI Commands
 
@@ -502,10 +526,15 @@ All commands are registered in [`src/gaia/cli.py`](src/gaia/cli.py). Run `gaia -
 - `gaia sd` - Stable Diffusion image generation
 - `gaia jira` - Jira integration
 - `gaia docker` - Docker management
+- `gaia browse` - Web research agent (search, page fetch, download)
+- `gaia analyze` - Structured data analysis agent (scratchpad tables)
+- `gaia email` - Email triage for Gmail (local inference; needs the Google connector)
 
 **Servers & infrastructure:**
 - `gaia api` - OpenAI-compatible API server
-- `gaia mcp {start|stop|status|test|agent|docker|serve|add|list|remove|tools|test-client}` - MCP bridge
+- `gaia mcp {start|stop|status|test|agent|docker|serve|list|tools|test-client}` - MCP bridge (add/remove moved to the connectors framework, #977)
+- `gaia telegram {start|stop|status}` - Telegram messaging adapter
+- `gaia connectors` - Manage connectors (Google/GitHub OAuth, MCP servers) and per-agent grants
 - `gaia cache {status|clear}` - Cache management
 
 **Setup & utilities:**
@@ -514,16 +543,16 @@ All commands are registered in [`src/gaia/cli.py`](src/gaia/cli.py). Run `gaia -
 - `gaia download` - Download a model
 - `gaia kill` - Kill stray GAIA / Lemonade processes
 - `gaia test` - Smoke tests
-- `gaia yt` - YouTube transcript ingest
-- `gaia template` - Scaffold agent templates
+- `gaia youtube --download-transcript <url>` - YouTube utilities (transcript download)
+- `gaia stats` - Show statistics from the most recent run
+- `gaia memory` - Manage agent memory (onboarding bootstrap, status)
+- `gaia diagnostics` - Bundle logs + system info into a tarball for bug reports
+- `gaia agent {export|import}` - Manage custom agent bundles
 
 **Evaluation & analysis** (see [`docs/reference/eval.mdx`](docs/reference/eval.mdx)):
-- `gaia eval {fix-code|agent}` - Run evaluation harness
-- `gaia gt` - Generate ground truth
-- `gaia generate` - Dataset/response generation
-- `gaia batch-exp` - Batch experiments
+- `gaia eval agent` - Run the agent eval benchmark (`--fix` auto-fixes failures)
 - `gaia report` - Render eval reports
-- `gaia visualize` / `gaia perf-vis` - Visualize results
+- `gaia perf-vis` - Visualize performance results
 
 **Standalone binaries** (separate `console_scripts`, not subcommands):
 - `gaia-code` - CodeAgent entry (`src/gaia/agents/code/cli.py`)
@@ -537,6 +566,9 @@ All documentation uses `.mdx` format (Markdown + JSX for Mintlify).
 **User Guides:**
 - [`docs/guides/chat.mdx`](docs/guides/chat.mdx) - Chat with RAG
 - [`docs/guides/agent-ui.mdx`](docs/guides/agent-ui.mdx) - Agent UI (desktop chat)
+- [`docs/guides/browse.mdx`](docs/guides/browse.mdx) - Web research (`gaia browse`)
+- [`docs/guides/analyze.mdx`](docs/guides/analyze.mdx) - Structured data analysis (`gaia analyze`)
+- [`docs/guides/email.mdx`](docs/guides/email.mdx) - Email triage (`gaia email`)
 - [`docs/guides/talk.mdx`](docs/guides/talk.mdx) - Voice interaction
 - [`docs/guides/code.mdx`](docs/guides/code.mdx) - Code generation
 - [`docs/guides/blender.mdx`](docs/guides/blender.mdx) - 3D automation
@@ -544,6 +576,12 @@ All documentation uses `.mdx` format (Markdown + JSX for Mintlify).
 - [`docs/guides/docker.mdx`](docs/guides/docker.mdx) - Docker management
 - [`docs/guides/routing.mdx`](docs/guides/routing.mdx) - Agent routing
 - [`docs/guides/emr.mdx`](docs/guides/emr.mdx) - Medical intake
+- [`docs/guides/telegram-adapter.mdx`](docs/guides/telegram-adapter.mdx) - Telegram messaging adapter
+- [`docs/guides/memory.mdx`](docs/guides/memory.mdx) - Agent memory
+- [`docs/guides/install.mdx`](docs/guides/install.mdx) - Installation
+- [`docs/guides/custom-agent.mdx`](docs/guides/custom-agent.mdx) - Build a custom agent
+- [`docs/guides/hardware-advisor.mdx`](docs/guides/hardware-advisor.mdx) - Hardware advisor
+- [`docs/guides/npu.mdx`](docs/guides/npu.mdx) - NPU setup
 
 **SDK Reference:**
 - [`docs/sdk/core/agent-system.mdx`](docs/sdk/core/agent-system.mdx) - Agent framework
@@ -587,13 +625,13 @@ The roadmap is at [`docs/roadmap.mdx`](docs/roadmap.mdx) ([live site](https://am
 - [`docs/plans/oem-bundling.mdx`](docs/plans/oem-bundling.mdx) - OEM hardware pre-configuration
 
 **Infrastructure:**
-- [`docs/plans/installer.mdx`](docs/plans/installer.mdx) - Desktop installer
-- [`docs/plans/mcp-client.mdx`](docs/plans/mcp-client.mdx) - MCP client integration
+- [`docs/plans/desktop-installer.mdx`](docs/plans/desktop-installer.mdx) - Desktop installer
+- [`docs/plans/mcp-docs.mdx`](docs/plans/mcp-docs.mdx) - MCP integration
 - [`docs/plans/cua.mdx`](docs/plans/cua.mdx) - Computer Use Agent
 - [`docs/plans/docker-containers.mdx`](docs/plans/docker-containers.mdx) - Docker deployment
 
 **Key architectural decisions (April 2026):**
-- ChatAgent renamed to **GaiaAgent** in v0.20.0 (#696)
+- **GaiaAgent** rename planned (#696) — not yet landed; the chat agent class is still `ChatAgent` (`src/gaia/agents/chat/agent.py`)
 - Voice-first is P0 enabling technology (#702)
 - No context compaction — memory + RAG handles long conversations
 - Configuration dashboard + Observability dashboard as separate Agent UI panels
@@ -614,7 +652,7 @@ When responding to GitHub issues and pull requests, follow these guidelines:
 
 The documentation is organized in [`docs/docs.json`](docs/docs.json) with the following structure:
 - **SDK**: `docs/sdk/` - Agent system, tools, core SDKs (chat, llm, rag, vlm, audio)
-- **User Guides** (`docs/guides/`): Feature-specific guides (chat, talk, code, blender, jira, docker, routing, emr)
+- **User Guides** (`docs/guides/`): Feature-specific guides (chat, browse, analyze, email, talk, code, blender, jira, docker, routing, emr, telegram-adapter, memory)
 - **Playbooks** (`docs/playbooks/`): Step-by-step tutorials for building agents
 - **SDK Reference** (`docs/sdk/`): Core concepts, SDKs, infrastructure, mixins, agents
 - **Specifications** (`docs/spec/`): Technical specs for all components
diff --git a/docs/guides/custom-agent.mdx b/docs/guides/custom-agent.mdx
index bb6c434ab..98a9cc9a9 100644
--- a/docs/guides/custom-agent.mdx
+++ b/docs/guides/custom-agent.mdx
@@ -138,7 +138,7 @@ If your agent needs a specific model, set `model_id` in `__init__()`:
 
 ```python
 def __init__(self, **kwargs):
-    kwargs.setdefault("model_id", "Qwen3-0.6B-GGUF")
+    kwargs.setdefault("model_id", "Gemma-4-E4B-it-GGUF")
     super().__init__(**kwargs)
 ```
 
diff --git a/docs/guides/mcp/client.mdx b/docs/guides/mcp/client.mdx
index be627b246..b4922c83a 100644
--- a/docs/guides/mcp/client.mdx
+++ b/docs/guides/mcp/client.mdx
@@ -3,6 +3,16 @@ title: "MCP Client"
 description: "Connect GAIA agents to external MCP servers and use their tools"
 ---
 
+<Warning>
+  **CLI update (#977):** `gaia mcp add` and `gaia mcp remove` shown in this guide
+  have been removed. MCP servers are now defined as **connectors** and configured
+  with `gaia connectors configure <id> --set KEY=VALUE` (secrets are stored in the
+  OS keyring; the entry is written to `~/.gaia/mcp_servers.json`). `gaia mcp list`
+  still lists configured servers. The `gaia mcp add ...` examples below predate this
+  change and are pending a rewrite — run `gaia connectors --help` for current usage,
+  and see issue #1339.
+</Warning>
+
 **Connect your GAIA agent to any tool ecosystem with a single line of code.**
 
 - 🔌 **Universal tool integration** - Filesystem, GitHub, databases, and hundreds more
diff --git a/docs/reference/cli.mdx b/docs/reference/cli.mdx
index 0f05d12bc..94440d054 100644
--- a/docs/reference/cli.mdx
+++ b/docs/reference/cli.mdx
@@ -880,6 +880,104 @@ gaia talk -i guide.pdf --no-tts
 
 ---
 
+### Jira Command
+
+<Card title="Jira / Atlassian" icon="ticket" href="/guides/jira">
+  Natural-language interface for Jira, Confluence, and Compass via the MCP bridge.
+</Card>
+
+```bash
+gaia jira "Create a bug report for the login issue"
+gaia jira --interactive
+```
+
+**Options:**
+
+| Option | Type | Default | Description |
+|--------|------|---------|-------------|
+| `command` | string | – | Natural-language command to execute (positional). |
+| `--interactive` | flag | false | Continuous interactive mode. |
+| `--mcp-host` | string | localhost | MCP bridge host. |
+| `--mcp-port` | integer | 8765 | MCP bridge port. |
+| `--verbose` | flag | false | Verbose output. |
+| `--debug` | flag | false | Debug logging. |
+
+[→ Full Jira Guide](/guides/jira)
+
+---
+
+### Docker Command
+
+<Card title="Docker" icon="docker" href="/guides/docker">
+  Natural-language interface for Docker containerization.
+</Card>
+
+```bash
+gaia docker "Create a Dockerfile for my Flask app"
+gaia docker "Containerize this project" --directory ./myapp
+```
+
+**Options:**
+
+| Option | Type | Default | Description |
+|--------|------|---------|-------------|
+| `command` | string | – | Natural-language command to execute (positional). |
+| `--directory` | path | current dir | Directory to analyze/containerize. |
+| `--verbose` | flag | false | Verbose output. |
+| `--debug` | flag | false | Debug logging. |
+
+[→ Full Docker Guide](/guides/docker)
+
+---
+
+### Summarize Command
+
+Summarize meeting transcripts, emails, and PDFs.
+
+```bash
+gaia summarize --input ./meeting.vtt
+gaia summarize --input ./inbox/ --type email --format both
+```
+
+**Options:**
+
+| Option | Type | Default | Description |
+|--------|------|---------|-------------|
+| `--input` | path | – | Input file or directory (required unless `--list-configs`). |
+| `--output` | path | auto | Output file/directory (auto-adjusted to format). |
+| `--type` | `transcript`\|`email`\|`pdf`\|`auto` | auto | Input type. |
+| `--format` | `json`\|`pdf`\|`email`\|`both` | json | Output format (`both` = json + pdf). |
+| `--styles` | one or more styles | executive participants action_items | `brief`, `detailed`, `bullets`, `executive`, `participants`, `action_items`, `all`. |
+
+---
+
+### Telegram Command
+
+<Card title="Telegram Adapter" icon="telegram" href="/guides/telegram-adapter">
+  Bridge a Telegram bot to GAIA so you can chat with your agents from Telegram.
+</Card>
+
+```bash
+gaia telegram start --token <BOT_TOKEN> --background
+gaia telegram status
+gaia telegram stop
+```
+
+**Subcommands & options:**
+
+| Subcommand | Option | Default | Description |
+|-----------|--------|---------|-------------|
+| `start` | `--token` (required) | – | Telegram bot token. |
+| `start` | `--allowed-users` | allow all | Comma-separated user IDs permitted to interact. |
+| `start` | `--background` | false | Run as a daemon (writes PID + health endpoint). |
+| `stop` | `--force` | false | Force stop if graceful shutdown fails. |
+| `status` | `--health-host` | 127.0.0.1 | Health server host. |
+| `status` | `--health-port` | 8765 | Health server port. |
+
+[→ Telegram Adapter Guide](/guides/telegram-adapter)
+
+---
+
 ## API Server
 
 <Card title="API Server" icon="server" href="/reference/api">
@@ -974,30 +1072,13 @@ Configure MCP servers that your agents can connect to. Servers are saved to `~/.
 
 ### Commands
 
-#### `gaia mcp add`
-
-Add an MCP server to configuration.
-
-```bash
-gaia mcp add <server-name> "<command>" [--config PATH]
-```
-
-**Arguments:**
-- `<server-name>` - Unique identifier for the server (e.g., "time", "memory")
-- `"<command>"` - Shell command to start the MCP server (must be quoted)
-
-**Options:**
-- `--config PATH` - Custom config file path (default: `~/.gaia/mcp_servers.json`)
-
-**Examples:**
-```bash
-# Add to user config (default)
-gaia mcp add time "uvx mcp-server-time"
-gaia mcp add memory "npx -y @modelcontextprotocol/server-memory"
+#### Managing MCP servers (`add` / `remove`)
 
-# Add to project config (can be committed to git)
-gaia mcp add time "uvx mcp-server-time" --config ./mcp_servers.json
-```
+<Note>
+`gaia mcp add` and `gaia mcp remove` were removed in #977 — MCP servers are now
+configured through the connectors framework. Run `gaia connectors --help` for the
+current commands. `gaia mcp list` (below) still lists configured servers.
+</Note>
 
 #### `gaia mcp list`
 
@@ -1019,29 +1100,6 @@ gaia mcp list
 gaia mcp list --config ./mcp_servers.json
 ```
 
-#### `gaia mcp remove`
-
-Remove an MCP server from configuration.
-
-```bash
-gaia mcp remove <server-name> [--config PATH]
-```
-
-**Arguments:**
-- `<server-name>` - Name of the server to remove
-
-**Options:**
-- `--config PATH` - Custom config file path (default: `~/.gaia/mcp_servers.json`)
-
-**Example:**
-```bash
-# Remove from user config
-gaia mcp remove time
-
-# Remove from project config
-gaia mcp remove memory --config ./mcp_servers.json
-```
-
 #### `gaia mcp tools`
 
 List tools available from a configured MCP server.
@@ -1258,35 +1316,14 @@ Use `lemonade-server list` to see all available models and their download status
 </Card>
 
 **Tools for:**
-- Ground Truth Generation
-- Automated Evaluation
-- Batch Experimentation
-- Performance Analysis
-- Transcript Testing
+- Agent eval benchmark (scenario-based, end-to-end)
+- Auto-fixing failures with Claude Code
+- Report generation
+- Performance-log visualization
 
 **Quick Examples:**
 
-Generate evaluation data:
-```bash
-gaia groundtruth -f ./data/document.html
-```
-
-Create sample experiment configuration:
-```bash
-gaia batch-experiment --create-sample-config experiments.json
-```
-
-Run systematic experiments:
-```bash
-gaia batch-experiment -c experiments.json -i ./data -o ./results
-```
-
-Evaluate results:
-```bash
-gaia eval -f ./results/experiment.json
-```
-
-Run agent eval benchmark:
+Run the agent eval benchmark:
 ```bash
 gaia eval agent                                    # Run all scenarios
 gaia eval agent --category mcp_reliability         # Run MCP reliability scenarios only
@@ -1296,14 +1333,14 @@ gaia eval agent --fix                              # Auto-fix failures with Clau
 gaia eval agent --compare run1/scorecard.json run2/scorecard.json  # Compare runs
 ```
 
-Generate report:
+Generate a report:
 ```bash
 gaia report -d ./eval_results
 ```
 
-Launch visualizer:
+Visualize llama.cpp performance logs:
 ```bash
-gaia visualize
+gaia perf-vis ./logs/llama-server.log
 ```
 
 [→ Full Evaluation Guide](/reference/eval)
@@ -1409,54 +1446,33 @@ gaia eval agent --scenario-dir ~/my-project/eval-scenarios
 
 ---
 
-### Visualize Command
+### Performance Visualization
 
-Launch interactive web-based visualizer for comparing evaluation results.
+Plot llama.cpp server performance metrics from one or more log files. Plots are saved as images; pass `--show` to also display them interactively.
 
 ```bash
-gaia visualize [OPTIONS]
+gaia perf-vis LOG_PATH [LOG_PATH ...] [--show]
 ```
 
 **Options:**
 
 | Option | Type | Default | Description |
 |--------|------|---------|-------------|
-| `--port` | integer | 3000 | Visualizer server port |
-| `--experiments-dir` | path | ./output/experiments | Experiments directory |
-| `--evaluations-dir` | path | ./output/evaluations | Evaluations directory |
-| `--workspace` | path | current directory | Base workspace directory |
-| `--no-browser` | flag | false | Don't auto-open browser |
-| `--host` | string | localhost | Host address |
+| `log_paths` | path(s) | - | One or more llama.cpp server log files to visualize |
+| `--show` | flag | false | Display plots interactively in addition to saving images |
 
 **Examples:**
 
 <CodeGroup>
-```bash Default
-gaia visualize
-```
-
-```bash Custom Directories
-gaia visualize \
-  --experiments-dir ./my_experiments \
-  --evaluations-dir ./my_evaluations
+```bash Single log
+gaia perf-vis ./logs/llama-server.log
 ```
 
-```bash Custom Port
-gaia visualize --port 8080 --no-browser
+```bash Multiple logs + show
+gaia perf-vis ./logs/run1.log ./logs/run2.log --show
 ```
 </CodeGroup>
 
-**Features:**
-- Interactive Comparison (side-by-side)
-- Key Metrics Dashboard
-- Quality Analysis
-- Real-time Updates
-- Responsive Design
-
-<Note>
-Node.js must be installed. Dependencies are automatically installed on first run.
-</Note>
-
 ---
 
 ## Memory
diff --git a/docs/spec/agent-base.mdx b/docs/spec/agent-base.mdx
index c6f98d979..d120118b1 100644
--- a/docs/spec/agent-base.mdx
+++ b/docs/spec/agent-base.mdx
@@ -119,7 +119,7 @@ def __init__(
         use_chatgpt: Use ChatGPT/OpenAI API instead of local LLM
         claude_model: Model to use when use_claude=True
         base_url: Local LLM server URL (default: from LEMONADE_BASE_URL or http://localhost:13305/api/v1)
-        model_id: Model ID for local LLM (default: Qwen3.5-35B-A3B-GGUF)
+        model_id: Model ID for local LLM (default: None — resolves to DEFAULT_MODEL_NAME, currently Gemma-4-E4B-it-GGUF)
         max_steps: Maximum reasoning iterations (default: 20)
         debug_prompts: Include prompts in conversation history (default: False)
         show_prompts: Display prompts sent to LLM (default: False)
diff --git a/docs/spec/autonomous-agent-mode.md b/docs/spec/autonomous-agent-mode.md
index e37cecc5a..a277932a4 100644
--- a/docs/spec/autonomous-agent-mode.md
+++ b/docs/spec/autonomous-agent-mode.md
@@ -952,7 +952,7 @@ async def _run_step(self, trigger: AgentTrigger, session: dict):
 **Settings:**
 - [ ] `agent_mode` in `SettingsResponse` + `SettingsUpdateRequest` (default `"autonomous"`)
 - [ ] GET/PUT `/api/settings` validates `agent_mode` against `"manual"|"goal_driven"|"autonomous"`
-- [ ] `GAIA_AUTO_OBSERVE_MODEL` env var wired into AgentLoop (default `"Qwen3-0.6B-GGUF"`)
+- [ ] `GAIA_AUTO_OBSERVE_MODEL` env var wired into AgentLoop (default `"Qwen3-4B-GGUF"`)
 
 **Audit Log:**
 - [ ] `GET /api/agent/activity` + `GET /api/agent/activity/{id}` endpoints
diff --git a/docs/spec/mcp-client.mdx b/docs/spec/mcp-client.mdx
index 00f214a30..4109659b1 100644
--- a/docs/spec/mcp-client.mdx
+++ b/docs/spec/mcp-client.mdx
@@ -4,6 +4,15 @@ description: "Complete technical specification for MCP client components"
 icon: "sitemap"
 ---
 
+<Warning>
+  **CLI update (#977):** `gaia mcp add` / `gaia mcp remove` referenced in this spec
+  were removed. MCP servers are now configured through the connectors framework
+  (`gaia connectors configure <id> --set KEY=VALUE`; the entry is persisted to
+  `~/.gaia/mcp_servers.json` with secrets in the OS keyring). `gaia mcp list` still
+  lists configured servers. Sections below using `gaia mcp add/remove` are pending a
+  rewrite — see issue #1339.
+</Warning>
+
 <Note>
 🔧 **You are viewing:** API Specification - Complete technical reference