Core Capability

CodePus Complete Feature Overview

Explore project development, Working, AI assistance, remote workspaces, agent collaboration, design, models, tools, memory, teams and billing.

Coding + Working
Project and everyday work
AI assistant
Your own conversation space
Agent collaboration
Plan, inspect and verify
Model choice
Cloud and personal connections

Coding, Working and AI across your devices

Develop software, handle everyday work and talk to your AI assistant from the workspace that fits your task.

Coding

Use the desktop IDE for code completion, project conversations, planning, implementation and debugging.

Download the IDE

Working

Continue work conversations on a connected computer, choose a project and manage tasks from the remote workspace.

Open Working remotely

AI assistant

Open a dedicated conversation workspace with text, supported voice options, personal memories and assistant appearance settings.

Open AI assistant

Remote

Connect to your authorized computers and continue Coding, Working or AI conversations. Desktop, terminal and file tools depend on the connected computer’s capabilities.

Open Remote

Sign in to open these workspaces. Remote requires an active paid plan and an online, authorized computer. Cloud AI uses your available Credits; supported features depend on your settings and device.

Explore the complete product

Choose a capability by the result you want. Each guide explains where to start and the conditions that apply.

Availability depends on the current client, model, account, permissions and configuration. The workspace shows which actions are available before you use them.

Core Capability

S0→S9 Full Lifecycle Pipeline

A staged pipeline for the software engineering lifecycle. The super agent advances under stage contracts with defined inputs, outputs and acceptance criteria. Final results are assessed against actual gate evidence.

Plan
00
S0

S0 Bootstrap

Task recognition · Type detection · Template matching

01
S1

S1 Requirements

Multi-source intake · Structured follow-ups · Multi-model debate validation

Design
02
S2

S2 Architecture

Stack selection · API design · Module breakdown

03
S3

S3 Design

URL→HTML · Image→Prototype · Voice→Mockups

Build
04
S4

S4 Environment

Auto-detection · Dependency install · Health check

05
S5

S5 Development

Task breakdown · Auto-coding · Change report

Verify
06
S6

S6 Testing

Tiered testing · Auto-diagnosis · Self-healing loop

07
S7

S7 Gate

Docs generation · Security scan · Quality gating

Operate
08
S8

S8 Release

Artifact packaging · Canary deploy · Delivery sign-off

09
S9

S9 Operations

Monitoring & alerting · Runbooks · Feedback iteration

One-Click Launch

Three quick-start templates covering every scenario

01

From Requirements to Delivery

S1→S9 fully automated, including architecture, environment, testing, gating, release and ops

02

Simulate a Website & Generate a Plan

S1→S5 with test loop, URL scraping + reverse requirements + automated implementation

03

Continue from an Existing Repo

S1→S5 recognises existing code and iterates incrementally

Core Engine

Super Agent · AI Engineering Commander

Beyond code completion, the agent can coordinate requirements, architecture, implementation and verification by stage, requesting human confirmation at key steps.

Full-Stack AI Commander

Commands 12 sub-agents to deliver complete software projects end-to-end. Covers requirements → architecture → design → coding → testing → release → ops.

Full-Lifecycle Orchestration

S0→S9 ten-stage orchestration supports skipping, parallel execution and rollback. Key actions remain subject to permissions and human confirmation.

Multi-Model Debate Engine

Use available provider and model configurations for cross-model checks to compare proposals and identify disagreements.

Self-Healing Loop

Test, diagnose, fix and retest within the configured round limit, preserving unresolved failures.

Quality-Gate Guardian

Each stage gated Pass/Warn/Fail — nothing passes that doesn't meet the bar

End-to-End Context

Requirements, architecture, design, code and test artifacts can pass between stages with verifiable records.

Human-in-the-Loop

Human confirmation at key checkpoints, autonomous decisions otherwise — efficiency and safety in parallel

Agent Army

12 Sub-Agents · End-to-End Expert Coverage

Every sub-agent is an expert in its niche — independent window, session, command and acceptance criteria. Launch individually or orchestrated by the super agent.

One-click orchestration · Staged delivery · Human confirmation at key steps

Plan & Design

Requirements Agent

S1

Multi-source intake (text/voice/URL/video/docs), structured follow-ups, multi-model debate, outputs PRD + acceptance criteria + feature-UI mapping

Output
PRD · Acceptance criteria · Feature mapping

Architecture Agent

S2

Tech stack selection, module decomposition, API list, data model design — auto-generates architecture draft

Output
Architecture draft · Tech stack · API list

Design Agent

S3

URL→HTML reverse engineering, image→prototype, sketch→code, voice-driven design edits, feature-UI mapping validation

Output
UI prototype · HTML code · Design mockups

Build & Implement

Config Agent

S4

Automated environment detection (doctor), dependency install (setup), runtime verification (verify) — ready in one click

Output
Environment ready · Dependency list · Health check

Coding Agent

S5

Reads requirements and design artifacts, decomposes the task tree, implements the code, emits change reports and risk notes

Output
Source code · Change report · Risk notes

Quality & Security

Test/DEBUG Agent

S6

Unit→integration→E2E→RPA tiered testing, auto-diagnosis + auto-fix + retest loop (configurable rounds)

Output
Test report · Coverage · Fix log

Docs Agent

S7

Auto-generates architecture docs, API docs, user manuals and ops guides — synced live with the code

Output
Architecture docs · API docs · Manuals

Security Agent

S7

Dependency scanning, risk grading (High/Medium/Low), fix suggestions, release-blocking decisions

Output
Security report · Risk matrix · Remediation plan

CI/CD Agent

S7

Pipeline execution, gate aggregation, artifact verification — automated CI/CD

Output
Pipeline result · Artifact bundle · Gate report

Ship & Operate

Release Agent

S8

Artifact packaging, release gates, canary deployment strategy, human confirmation for production, delivery manifest

Output
Release package · Canary strategy · Delivery manifest

Ops Agent

S9

Monitoring & alert rules, Runbook generation, incident response flow, database/server deployment config

Output
Runbooks · Monitoring rules · Ops manuals

Feedback & Iteration Agent

S9+

Collects issues/user feedback, prioritises them, one-click loops back to S1 to start the next iteration

Output
Feedback triage · Priority list · Iteration plan
Command Center

Agent Hub · Global Project Cockpit

Real-time visibility into every agent's status, stage progress and quality-gate results. When the super agent is running, you don't see a black box — you see a full stage timeline, progress bars and deliverable list.

Stage Timeline

S0→S9 visual timeline, current stage highlighted, progress at a glance

Runtime Metrics

Success rate, average duration, failure point tracking — every delivery quantified

Quality Gates

Pass/Warn/Fail three-tier gating — nothing ships until it qualifies

Deliverables Cockpit

Requirements, design prototypes, test reports, delivery lists — all in one view

Self-Healing Loop

Tests Failed? AI Auto-Fixes Them

Debug mode runs tests, diagnoses failures, applies fixes and retests within the current round, permission and resource limits. Unresolved failures and diagnostics remain available when a limit is reached.

1

Run Tests

Unit · Integration · E2E · RPA tiered execution

2

Auto-Diagnose

Aggregate logs, traces, screenshots to pinpoint root cause

3

Auto-Fix

Automatically apply fix patches based on diagnosis

4

Retest & Verify

Auto-retest after fix to confirm resolution

Multi-Modal Input

Beyond Typing — Speak, Screenshot, Paste a Link

CodePus supports multi-modal input across chat and every working mode. Express your requirements the most natural way.

Text
Describe requirements and intent in natural language
Voice
Record → Transcribe → Edit → Confirm, ASR supports multiple languages
URL
Auto-scrape web pages, configurable depth (0=full site)
Image
Screenshot→HTML, Sketch→Prototype, Image→Code
Video
Multimodal video parsing to extract requirements and design
Document
PDF/DOC/MD auto-parsed into structured requirements
Collective Intelligence

Multi-Model Debate, Collective Decision

Got a complex requirement? Multiple AI models debate the contested points and output a converged plan. Not one model's guess — multiple models colliding into the best answer.

1

At least 2 different models join the debate

2

User can stop anytime and take the latest consensus doc

3

Disagreements → merged conclusions → open questions all logged

Ultimate Experience

AI Programming Core Capabilities

From millisecond completions to deep code understanding, every detail engineered for peak efficiency.

Tab Smart Completion

Proprietary completion model with millisecond response, multi-line prediction, function-level generation. Context-aware precision for your next move.

Ctrl+K Inline Edit

Select code, describe changes in natural language, AI rewrites instantly. Diff preview + AI review — see before you change.

@codebase Deep Search

Semantic indexing of your entire codebase. Search code in natural language, understand architecture intent beyond text matching.

Intelligent Agent Chat

Built-in AI Agent understands your entire codebase, supports multi-turn conversation, @file precise references, full context continuity.

Background Agent

Background agents take on complex tasks — multi-file refactors, test generation, bug hunts — without making you wait.

BugBot PR Review

AI auto-reviews pull requests, flags potential bugs, performance issues and security risks, and proposes fixes.

Model Capabilities

7 Capability Dimensions · Smart Model Routing

Choose from currently available models or Auto, and configure supported local models or your own provider connections. Available capabilities depend on your current setup.

Reasoning

Deep Reasoning

Complex logic · Math proofs · Multi-step planning

Coding

Code Generation

Smart coding · Refactoring · Auto testing

Multimodal

Multimodal Understanding

Image+text · UI screenshot analysis · Visual reasoning

LongCtx

Long Context

Million-token windows · Full repo awareness

Speed

Blazing Speed

Millisecond inference · Real-time completion · Streaming

Privacy

Privacy First

Local deployment · Data stays on-premise · Compliance

Custom

Custom Integration

OpenAI-compatible · Private deploy · Self-hosted models

Tools & Ecosystem

Infinitely Extensible AI Capabilities

Through MCP protocol, project rules, and privacy mode, CodePus's AI capabilities extend without limits.

MCP Protocol Support

Connect databases, APIs, docs and external tools via Model Context Protocol for infinite AI extension. Built-in MCP marketplace, one-click install.

Project Rules (.codepus/rules)

Define coding standards and tech stack constraints via project-level rule files, ensuring AI-generated code meets team standards. Version-controlled, team-shared.

Privacy Mode

Local models can be configured for sensitive projects. Actual data paths depend on the models, tools and deployment you enable.

Terminal Integration

AI reads terminal output, understands build errors, automatically provides fix solutions and executes them directly.

Smart Refactoring

AI understands code intent, auto-suggests refactoring plans. Supports function extraction, variable renaming, structure optimization.

Context References

Use @ to reference files, folders, code symbols, doc links — let AI precisely understand your intent.

Team Collaboration

Built for Teams

Share prompts, rules, and tool configs for a unified team AI coding experience.

Team Spaces

Share prompts, rules, and tool configs. Unified team AI coding experience — new members onboard instantly.

Usage Analytics

Per-member AI usage, model consumption, and code generation stats to optimize team efficiency.

RBAC Access Control

Fine-grained role-based access control with admin, member, and viewer tier management.

Security & Compliance

Enterprise Security. Zero Compromise.

Your code is your most valuable asset. CodePus provides comprehensive security so enterprises can confidently adopt AI coding.

Privacy Mode

Local models can be configured for sensitive projects. Actual data paths depend on the models, tools and deployment you enable.

SSO Single Sign-On

SAML/OIDC enterprise auth, SCIM auto-sync seats, unified identity management.

Audit Logs

Operation auditing is available. Certification and compliance status depend on formal third-party audit results.

Private Deployment

Private deployment can be configured. Data boundaries depend on your actual model, tool and infrastructure settings.

The Future of Coding Is Here

Choose your workspace, read the relevant guide and start with the client and capabilities available to you.