Archietect logo

Archietect

Free

Deterministic architectural memory engine — answers whether a concept already exists, what's canonical, and what the evidence is, before an AI or a human builds it again.

Visit Website
Related by shared tags and keywordsai@saroday/mcp-server@globestudio/mcp
ai-toolsclideveloper-toolsmcpmodel-context-protocolaiai assistantautomation

Overview

Archietect is a deterministic engine that maintains a living record of what a project's concepts are, including what exists, what is canonical, who uses it, why it is shaped this way, and how that has changed over time. It operates behind an explicit permission boundary controlling what it is allowed to look at, and includes a registry that knows about every project it has been pointed at on the machine, not just the one open right now. AI agents, editors, CI, and humans all query the same continuously maintained state instead of each reconstructing a partial, inconsistent understanding per session. Archietect does not use AI; AI uses Archietect. It provides evidence-tiered answers ranked as DECLARED, USED, or NAMED, and is read-only against code, flagging problems with evidence and stopping without editing the working tree. It is fully offline and deterministic, with no AI inside, no API key, and no network call. The system supports three transports—CLI, REST, and MCP—all calling the same query functions, and uses Rust, SQLite, regex-based structural/schema extraction, tiny_http for REST, and stdio for MCP. Query latency for read-only commands is approximately 55ms on a 637MB / 30k-file repository, and context efficiency on real repositories is consistently 98–99% discovery payload reduction.

Pricing

Free

Detailed plans have not been confirmed in our catalog. Check the official website for current limits and billing terms.

Visit Website

Prices and limits may change. Confirm the currency, billing period, seat minimum and usage caps on the official website.

Use Cases

  • Teams use Archietect to check if a concept already exists in the codebase, where it is, and how confident the system is.
  • Developers use Archietect to get a git-status-style glance at what it found in a project.
  • Users use Archietect to view the same information in a browser without remembering commands.
  • Teams use Archietect to trace downstream dependencies of a dbt model and see every mart and dashboard that depends on it.
  • Developers use Archietect to trace cross-silo linkages from a backend ORM model into dbt models materializing the same table.
  • Teams use Archietect to detect duplicate business logic across concepts.
  • Users use Archietect to check context efficiency and see the median source bytes an agent would read versus the JSON receipt returned.

Features

AI assistant

Natural language interface

Workflow automation

Reviews

No reviews yet. Be the first to share your experience!

Rankings & collections