What tools does Safer Agentic AI MCP provide?
It provides 12 read‑only tools for searching patterns, resolving requirements, mapping tasks to safety guidance, and following cross‑references.
Safer Agentic AI MCP Server
@NellInc
Basic information
About Safer Agentic AI MCP Server
SaferAgenticAI MCP brings the Safer Agentic AI safety framework to coding assistants over the Model Context Protocol.
Safer Agentic AI MCP Server config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"saferagenticai": {
"type": "http",
"url": "https://mcp.saferagenticai.org/mcp"
}
}
}Safer Agentic AI MCP Server tools
12List all 16 suites in the SaferAgenticAI framework (9 drivers + 7 inhibitors) with subgoal counts and titles. Call this first to orient.
Retrieve one subgoal (framework normative content + Pattern layer guidance) by pattern_id (e.g., 'D3::idx2::sandboxing') or display_id (e.g., 'D3.2'). display_id may resolve to multiple subgoals — underlined variants share display_ids.
List subgoals matching filters (suite_id, suite_type, content_type, min_confidence, missing_pattern_only). Results capped by limit (default 50, max 100).
Field-weighted keyword search across the framework. Substring match on lowercased terms; field weights: title 10x, summary 4x, SFR text 3x, description 2x, pattern body 1x. `matched_in` reports the highest-weighted field that matched. No semantic / embedding search — known limitation, see /mcp.html. Use verbosity='compact' to drop snippets and confidence flags (~70% smaller payload) when triaging.
Return outgoing adjacencies for a pattern. `explicit_cross_references` are author-asserted (each pattern's `cross_references` YAML field). `inferred_adjacent` (when include_inferred=true) currently returns *same-suite siblings only* — it does not do semantic similarity. Treat inferred entries as 'neighbours worth scanning,' not as endorsed dependencies.
Resolve a loose reference (partial id, display_id, slug fragment, or title keyword) to canonical pattern_id(s). Call this when you have a rough reference and need the exact id before calling get_requirement. Always returns candidates — never 'not found'.
Given a natural-language task description (e.g., 'I'm building a tool-using agent that runs shell commands'), return the most relevant patterns grouped by suite. Use this as a starting point for any cross-cutting design question; then follow up with get_requirement on specific pattern_ids. Defaults to verbosity='compact' (cheap triage); pass 'full' to inline snippets and confidence flags.
Return patterns that have not been human-reviewed yet (no reviewed_by). Sorted low-confidence first, then needs_human_review flagged, then alpha. Use during Phase 3 review to pick the next pattern to examine.
Coverage stats: total patterns, reviewed %, per-suite and per-confidence breakdown. Surfaces load-time validation issue count.
Return patterns that reference the given pattern_id in their cross_references. Complement to get_cross_references (outgoing); this shows incoming. Use to find all consumers of a given pattern.
List operational heuristics distilled from production agentic AI deployment (Claude Code, Rewind). These are cross-cutting safety principles discovered through building and operating AI agents, mapped to framework suites. Optional filters: suite_id (heuristics relevant to a specific suite), query (keyword search across titles and principles). Separate from the normative pattern layer — different category of knowledge.
Retrieve a single operational heuristic by id (e.g., 'OH::geoffrey-pattern'). Returns the full entry: principle, framework mapping, evidence sources from production deployment, design patterns, anti-patterns, and discovery narrative.
Overview
What is Safer Agentic AI MCP?
Safer Agentic AI MCP brings the Safer Agentic AI safety framework to coding assistants over the Model Context Protocol. It serves 238 normative implementation patterns and 14 operational heuristics across 16 safety suites (9 Drivers + 7 Inhibitors) through 12 read-only tools.
How to use Safer Agentic AI MCP?
Install the package with uvx saferagenticai-mcp or pipx install saferagenticai-mcp, then add a configuration entry to your MCP client. The tools allow you to search patterns, resolve requirements, map a task to relevant safety guidance, and follow cross‑references.
Key features of Safer Agentic AI MCP
- 12 read‑only tools for safety guidance access
- 238 normative implementation patterns and 14 operational heuristics
- Covers all 16 safety suites (9 Drivers + 7 Inhibitors)
- No authentication or external network calls required
- All framework data is bundled in the package
- stdio transport, Python ≥3.10, MIT licensed
Use cases of Safer Agentic AI MCP
- Check code against normative safety patterns during development
- Resolve specific safety requirements for an AI agent task
- Map a user’s task to the relevant safety guidance from the framework
- Follow cross‑references between patterns and heuristics
- Train or assist coding assistants with structured safety knowledge
Frequently asked questions
What are the system requirements?
Python 3.10 or higher is required, and the server can be installed via `uvx` or `pipx`.
Does Safer Agentic AI MCP make external network calls?
No. All framework data is bundled locally, and no authentication or external calls are needed.
What transport protocol is used?
The server uses stdio transport.
What license is the project under?
MIT.
More Reasoning MCP servers
ArduPilot MCP Server Sandbox
hfujikawa77ArduPilotドローンをAIエージェントから操作するMCPサーバーです。
9
MCP Servers
junfanz1This project demonstrates a decoupled real-time agent architecture that connects LangGraph agents to remote tools served by custom MCP (Modular Command Protocol) servers. The architecture enables a flexible and scalable multi-agent system where each tool can be hosted independent
27
Code Reasoning MCP Server
mettamattA code reasoning MCP server, a fork of sequential-thinking
255
IntelliNode Medical Use Cases
BarqawizMulti-Agent AI Orchestration Workshop
9
MCP Sandbox
JohanLi233Python sandboxes for llms
40
Comments