asdecided-core
Provides deterministic, read-only access to engineering decisions, requirements, and designs stored as Markdown in a repository, enabling agents to retrieve and follow decided context without embeddings or external services.
README
AsDecided
Engineering decisions your agents can follow. Build, as decided.
AsDecided keeps requirements, decisions, designs, roadmaps, and prompts as typed Markdown in your repository. Its native Rust engine validates that knowledge, retrieves relevant decisions deterministically, and serves it read-only to agents over MCP.
No embeddings, model call, hosted index, or Python runtime is required. The same repository state produces the same answer.
Install
Install the native engine and MCP server through Homebrew:
brew install asdecided/tap/asdecided-core
Rust users can install the decided CLI directly from crates.io:
cargo install decided
cargo install decided-mcp
Windows users can install both native executables through Scoop:
scoop bucket add asdecided https://github.com/asdecided/scoop-bucket
scoop install asdecided
Native decided and decided-mcp archives are also published on
GitHub Releases.
rac-core is no longer distributed through PyPI. Python API consumers should
use asdecided/sdk, which is a
client SDK rather than a second engine implementation.
Start a repository
decided quickstart
decided validate decisions/
decided gate decisions/
decided gate decisions/ --code --base origin/main
New repositories use:
.decided/config.yaml
decisions/
Existing artifact IDs such as RAC-ABC123DEF456 are durable identities and do
not change with the product name.
Migrate an existing repository
Migration is explicit and never runs during an ordinary command:
decided migrate layout . --dry-run
decided migrate layout .
The migration moves .rac/ to .decided/ and rac/ to decisions/. It
refuses to overwrite either destination.
MCP
The official MCP Registry identity is io.github.asdecided/core.
{
"mcpServers": {
"asdecided": {
"command": "decided-mcp",
"args": ["--root", "."]
}
}
}
Architecture
Rust is the product engine and the only CLI/MCP runtime in this repository.
The authoritative language-neutral compatibility fixtures live in
asdecided-spec. Live-corpus validation is
based on validity, determinism, freshness, and cache/no-cache equality.
Document ingestion remains an ancillary Python connector rather than part of
the core engine. The retired Python engine is preserved for historical review
at the immutable
python-engine-final
tag; it is not maintained or run in normal CI.
License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.