Features

Autopilot Cruise · Basic

Type a goal in the prompt box, press flight PLAN, and Cockpit takes the yoke:

• A planner turns the goal into steps and assigns the best agent/model to each
• It runs as a durable, resumable chain — checkpointed, auto-recovering if interrupted
• The answer lands back in your chat; you clear only the one critical moment
• Fast-path for trivial messages (skips planning) · Enter always dispatches, flight PLAN is the only thing that plans

AutoCognition · Pro

A shared local brain your agents orchestrate themselves on:

• hive Agent bus — presence; agents see, message, and call each other
• Durable task chains + board · granularity · checkpoints · auto-resume-on-kill
• Capability routing (best-fit agent) · quorum voting (models vote, a judge decides)
• Consent-gated broker · approvals queue · turn any Power Combo into a durable chain

Five Supplier Classes

• Card 1 — Agent CLI — default Claude Code; or pick a free agent (Aider, Goose, opencode) or another CLI
• Card 2 — Agent CLI — default Codex; run it alongside Card 1, or drop in a free agent / Gemini (free API key)
• Card 3 — Local LLMs — Ollama, LocalAI, or Dual — Ollama + LocalAI (both at once, for regular chat too, not just Power Combos). Other local servers (LM Studio, llama.cpp) can be added as a custom Worker instead
• Workers — Cloud API providers (Groq, OpenRouter, Gemini API)
• Tools — Local CLI binaries (ffmpeg, whisper, pandoc)

Every slot is a configurable pick (settings Configure supplier slots) with a [Free] flag on the no-cost options, and a one-click Stay Free that fills the cards with free agents on your local models. The app detects your real GPU/VRAM (AMD + NVIDIA) so a Broadcast to two local agents that won't co-fit runs them one at a time instead of thrashing.

Multi-Agent Modes

• Single — active agent only
• Pipeline — chain A→B→C
• Broadcast — parallel to all
• Plan — voting orchestrator

Streaming Responses

200ms polling, file-queue architecture. No WebSocket server, no persistent process. Survives Apache restarts. Works on standard shared Apache + PHP hosting.

Plan Mode / Orchestrator

All three agents propose plans independently. A synthesizer merges the votes. Execute step by step, with skip and edit per step. Four auto-run modes.

%%MARKER%% Protocol

Agents write structured markers in their output. %%CREATE%% writes files, %%WORKER:%% calls cloud APIs, %%TOOL:%% runs local binaries, %%PROPOSE:%% creates review proposals.

Context per Project (COCKPIT.md)

Place a COCKPIT.md file in any project folder. Its contents are automatically prepended to every agent call — no more repeating conventions and file locations every session.

Parallel

Several small models answer the same prompt independently — good for comparing takes or brainstorming, fast since they run concurrently.

Sequential

Output of one model feeds the next as a pipeline — draft → critique → refine, all in one click.

Judge

Several candidates get generated, then a synthesizer model picks or merges the best answer. Optional shared-premise seeding keeps parallel creative combos on the same idea instead of diverging randomly.

Set it once, forget it

Define rules that match keywords or phrases in your prompt and automatically hand that task to a specific worker, combo, or Ollama model — regardless of which mode (Single, Pipeline, Broadcast, Plan) is currently active.

Smart match cascade

Keyword match first (instant, free). Ambiguous cases fall back to a small local classifier model. Still unclear? Cockpit asks you — it never silently guesses on a rule that doesn't clearly apply.

Transparent hand-off

The delegated result is spliced back into your conversation and the original agent continues — you see exactly what got delegated and what came back, nothing happens invisibly.

Style Strip

Switch prompt modes instantly: Chat (conversational), Plan, Research, Build, Debug, Analyze, Report (HTML output). Style saved per chat, persists across reloads.

Prompt History

Shell-style ↑/↓ navigation through your last 50 prompts, per chat. Live draft preserved when navigating. Only triggers on the first/last line — multi-line editing works normally.

Proposals System

Agents propose code snippets using %%PROPOSE:snippet%%. Review each proposal, approve to global library or current project, or reject. Tolerates small model syntax variations.

Session Summary

One-click session summary: asks the active agent to produce a structured HTML summary of the conversation, saved to the chat's context folder for future reference.

Translation Button

On-demand translation of selected agent output text using local LibreTranslate. English, Spanish, Hebrew. No automatic UI translation — you translate specific text when you need it.

Chat Export / Import

Export any chat as a ZIP (JSON + markdown transcript + manifest). Import from ZIP with automatic name conflict resolution.

Smart Attachment Routing

• Text/code → inlined directly
• PDF/DOCX/EPUB → pandoc → markdown
• Audio → whisper.cpp → transcript
• Video → ffmpeg extract → whisper → transcript
• Images → vision model → description
• Archives → saved, path reference

Image Paste

Paste an image from clipboard directly into the prompt textarea. It is automatically uploaded and routed through the same vision pipeline as file attachments.

Per-Chat Storage

Attachments saved to chat_<name>_contexts/ in the project folder. Cleaned up when the chat is deleted. Viewable from the Attachments panel in the right sidebar.

Keyword RAG Search

Instant offline search over your local snippets, knowledge, and library items. Top 3 results offered as Pending Proposals — never silently injected (that wastes tokens).

Semantic Deep Search

Optional "Deep Search" button uses nomic-embed-text via Ollama to find conceptually similar items even when keywords don't match. Costs one Ollama call per search.

Accumulator

Feed any PDF, DOCX, EPUB, HTML, TXT, or CHM file. Extracts code snippets and knowledge items as structured JSON for RAG retrieval. Schedule overnight to avoid blocking the UI.

Internal Notes (RAG-linked)

A dedicated left-panel Notes tab, separate from Library/Snippets — its own searchable store, optionally cross-linked into the same RAG context scope so notes and code snippets surface together.

Notepad (folder mode)

Point Notepad at any local folder on your machine and it becomes a second, independent notes store — for people who already keep notes the OpenNote way and don't want a walled-garden format.

Cloud Workers

Groq, OpenRouter, HuggingFace, Together AI, DeepInfra. API keys encrypted at rest (AES-256-GCM, machine-bound). Per-provider rate-limit tracking with daily quota bars.

Local CLI Tools

ffmpeg (9 presets), whisper.cpp (7 presets), pandoc (8 presets). Run from within Cockpit or triggered automatically by agents via %%TOOL:%% markers. Output inlined or path-referenced.

Web Search

Brave Search API primary, DuckDuckGo fallback chain. Results shown in the Search panel and optionally injected as context into the next prompt.

Translate chat / text

Translate the whole conversation or a selected snippet on demand, via local LibreTranslate — no automatic UI translation, only what you ask for.

Skills

Reusable prompt/instruction bundles an agent can load mid-conversation via a marker, instead of you re-typing the same context every time.

Delegate

Opens the task auto-routing rule manager — see above.

Track

Pin a running thread of decisions/TODOs alongside the chat, so long sessions don't lose the thread.

Summarize

One click, structured HTML summary of the current session, saved for later reference.

Image / Video / Sound

Direct access to local media-generation and processing tools without leaving the chat.

Terminal

A real terminal panel into your own machine — same server the agents run on, for when you want to do something yourself.

Files / Browser

Files: a project-scoped file browser/picker. Browser: an embedded web view for quick lookups without tabbing away.

Keys

Manage your own cloud-provider API keys, encrypted at rest, machine-bound — never sent anywhere but the provider you're calling.

Tasks

A lightweight task list panel, separate from chat history, for tracking what still needs doing across a project.

Notepad

See RAG & Knowledge Base above — folder-backed notes independent of the internal Notes store.

36 Themes

14 dark + 13 light + 9 mixed (dark chrome + light console). Includes Dracula, Nord, Monokai, One Dark, Gruvbox, Tokyo Night, Solarized, Catppuccin, and more.

18 Protanopia-Safe Dalton Themes

Designed for red-green color deficiency. Uses blue, yellow, cyan, and orange exclusively — never red or green as the sole distinguishing factor. Auto-contrast safety net ensures readability on every theme.

Real Usage Tracking

Every supplier card's token counts (Cards 1–3 plus Workers) read directly from their log files. Per-day / week / month / all-time ranges. Daily quota bars for cloud providers.