OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Coding Agents

Invincat

A native Python terminal agent for AI-assisted work in local repositories, with a separate memory agent and hierarchical memory management.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Invincat — OpenViST project profile in Coding Agents.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Invincat when repository work is best handled from a terminal and previous project decisions or conventions should remain available in later sessions. Use `/plan` for risky, multi-file, ambiguous or architecture-level work that needs review first.

Who it is for

  • Developers working in local repositories from a CLI.
  • Users who want coding assistance that retains project-specific learnings.
  • Developers who need approval before implementation changes.

Documented uses

  • Inspect, create and edit repository files.
  • Run approved shell commands during coding tasks.
  • Review an execution plan before implementation.
  • Maintain user and project memory across sessions.
  • Pursue a long-running objective across multiple turns.
Before you choose

Requirements & limits

Invincat is documented as a Python CLI and requires Python 3.11 or later. The public repository displays an MIT license badge. First launch requires configuring and activating a model through `/model` or documented environment variables.

  • The tool works from a local project directory rather than a browser interface.
  • File changes and shell commands use approval-gated execution, while plan mode separates review from implementation.
  • Memory is split into user and project scopes, with project entries stored under .invincat/memory_project.json.
  • A dedicated memory model is optional; without one, extraction uses the current primary model.
  • MCP tools, skills, scheduled tasks, goal mode and the documented WeCom integration extend the basic coding-agent workflow.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install and start Invincat in a project

Set up the CLI and open it from a local project directory.

Before you start

  • Python 3.11 or later.
  • A local project directory.
  • A model provider, model name and API key for model configuration.

Steps

  1. Install the package with the command below.

    Documented command
    pip install invincat-cli

  2. Change to the project directory.

  3. Run invincat-cli.

  4. On first launch, run /model and register a provider, model name, API key and optional base URL.

  5. Select the configured model and press Enter to activate it.

Expected outcome

Invincat opens in the project directory with an activated model for coding assistance.

Guide limitations

  • The documented setup does not establish provider pricing, model availability or compatibility beyond the stated Python requirement.
  • The primary service action requires model configuration after first launch.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
30 September 2026
Original website title
dog-qiuqiu/invincat · Official homepage title · retained in its source language
How we review sourcesSuggest a correction