gjalla

Official registry AI and agents

Hosted self-curating shared memory that keeps your agents working like a high-performing team

Install

Add this to your client's MCP configuration (Claude Desktop: claude_desktop_config.json; Cursor: .cursor/mcp.json; VS Code and others use the same "mcpServers" shape).

Remote (streamable-http)
Claude Code
claude mcp add --transport http gjalla 'https://gjalla.io/mcp/stream' --header 'Authorization: <Bearer <gjalla agent API key>. Generate one at https://gjalla.io/account/api-keys.>'
JSON configuration (Claude Desktop, Cursor, VS Code)
{
    "mcpServers": {
        "gjalla": {
            "type": "http",
            "url": "https://gjalla.io/mcp/stream",
            "headers": {
                "Authorization": "<Bearer <gjalla agent API key>. Generate one at https://gjalla.io/account/api-keys.>"
            }
        }
    }
}

Documentation

From the project's README on GitHub, refreshed weekly. Links open on GitHub.

gjalla/engineering

Opinionated and time-tested agent skills for trustworthy, production-grade engineering — spec-driven planning, rigorous review, test auditing to bust false-confidence, and clean operations. Built with engineering best practices in mind, adapted for use by teams shipping with coding agents. Welcome to agentic SDLC!

Install

# all skills
npx skills add gjalla/engineering

# browse and pick
npx skills add gjalla/engineering --list

Then tell your agent "set up gjalla". The gjalla-onboard skill installs the CLI, walks you through one browser sign-in, backfills your last week of agent sessions, and reports what your coding-agent spend bought by task type. Alternatively start from the CLI with uvx gjalla setup, which installs these skills for you.

Each skill is a SKILL.md your coding agent loads on demand. Works with Claude Code, Cursor, Codex, and the other agents supported by npx skills.

Every skill works standalone. A few note optional gjalla CLI commands that enrich them with live architecture and rules context — those are enhancements, not requirements.


Built by gjalla — the infrastructure for trustworthy code when building with coding agents.

Frequently asked questions

What is gjalla?

Hosted self-curating shared memory that keeps your agents working like a high-performing team

How do I install gjalla?

It runs as a remote server. In Claude Code: claude mcp add --transport http gjalla 'https://gjalla.io/mcp/stream' --header 'Authorization: <Bearer <gjalla agent API key>. Generate one at https://gjalla.io/account/api-keys.>'. For other clients, paste the JSON configuration from this page into their MCP settings.

Does gjalla need an API key?

Yes. It expects Authorization, which you set in its configuration.

Is gjalla open source?

Its code is public on GitHub, but the repository does not declare a standard license.

Who maintains gjalla?

It is published by gjalla. AI.tn lists it but does not maintain or endorse it.

From the official MCP Registry. Listing is not an endorsement: review a server's code and permissions before giving it access to your data.