io.github.NellInc/saferagenticai-mcp

Measured conclusion: Deep-tested. L2 outcome: tools_ok.

Read-only tools over the Safer Agentic AI framework: 238 patterns + 14 heuristics.

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "saferagenticai-mcp": {
      "url": "https://mcp.saferagenticai.org/mcp",
      "transport": "streamable-http"
    }
  }
}

Measured tools (10)

• list_suites read-only

  List all 16 suites in the SaferAgenticAI framework (9 drivers + 7 inhibitors) with subgoal counts and titles. Call this first to orient.

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• get_requirement read-only

  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.

  Parameters:

  • id (required) — string
  • include_pattern — boolean

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• list_requirements read-only

  List subgoals matching filters (suite_id, suite_type, content_type, min_confidence, missing_pattern_only). Results capped by limit (default 50, max 100).

  Parameters:

  • limit — integer
  • suite_id — string
  • suite_type — string
  • content_type — string
  • min_confidence — string
  • include_pattern — boolean
  • missing_pattern_only — boolean

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• search_patterns read-only

  Field-weighted keyword search across the framework. Terms match at word starts on lowercased text and are IDF-weighted, so rare terms outrank ubiquitous ones; field weights: title 10x, summary 4x, SFR text 3x (all of a subgoal's SFRs scored as one field), 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.

  Parameters:

  • query (required) — string
  • limit — integer
  • verbosity — string

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• get_cross_references read-only

  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.

  Parameters:

  • id (required) — string
  • include_inferred — boolean

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• resolve_id read-only

  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'.

  Parameters:

  • query (required) — string
  • limit — integer

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• find_patterns_for_task read-only

  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.

  Parameters:

  • task (required) — string
  • limit — integer
  • verbosity — string

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• get_reverse_references read-only

  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.

  Parameters:

  • id (required) — string

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• list_operational_heuristics read-only

  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.

  Parameters:

  • query — string — Keyword search across titles, principles, narratives
  • suite_id — string — Filter by framework suite (e.g., 'D3', 'I2')

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• get_operational_heuristic read-only

  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.

  Parameters:

  • id (required) — string

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

Repository

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.