NeuroDock Guardrail

io.github.tlennon-iePublished by GitHub account tlennon-iev0.0.5Updated Oct 24, 2026

Advisory detectors for rumination, hyperfocus, and sycophancy. Never silently blocks.

VerifiedSTDIODesktop onlySecurity & MonitoringAI & ML

Overview

AI-generated overview

Advisory detectors that flag rumination loops, unregulated hyperfocus, and sycophancy in assistant conversations without blocking them.

What you can do
NeuroDock Guardrail exposes three advisory tools: check_rumination, which uses a word-overlap Jaccard heuristic over a caller-supplied history window; check_hyperfocus, which uses an elapsed-time threshold with end-of-day handling; and check_sycophancy, which matches four validation patterns. Each returns a detection result with default thresholds that callers can adjust. The server is stateless, persists nothing, and never silently blocks a conversation.
When to use it
Worth adding when an assistant is used by or with neurodivergent people and you want a lightweight, auditable signal for repetitive loops, runaway focus sessions, or excessive agreement. It is advisory only, so it suits setups where the assistant or user decides what to do with a flag rather than an automatic gate.
Requirements
Runs locally as a stdio process, installed from PyPI as neurodock-mcp-guardrail and started with uvx or its console script. No accounts, API keys, environment variables, headers, or network access are declared. Callers must supply conversation history themselves, since the server stores nothing.
Before you install
Detections are heuristic and threshold-based, so false positives and false negatives are expected; treat flags as suggestions, not clinical judgments. The server is stateless and keeps no history, so results depend entirely on what the caller passes in. It declares no telemetry and no network sockets, and logs only tool-invocation metadata.

Installation

In SourceWeft

  1. Open NeuroDock Guardrail in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

neurodock-mcp-guardrail

Clinical guardrails MCP server for NeuroDock. Detects three patterns that unmodified LLM interaction tends to amplify for neurodivergent users: rumination loops (OCD), unregulated hyperfocus (ADHD), and sycophancy / over-validation. See ETHICS.md and MANIFESTO.md for the full framework.

Status

  • v0.0.2: All three detectors are live.

    • check_rumination — word-overlap Jaccard heuristic.
    • check_hyperfocus — elapsed-threshold-with-end-of-day heuristic.
    • check_sycophancy — four-pattern overlap heuristic.

    Heuristics are public and auditable; thresholds are defaults, not prescriptions. See ADR 0006 for the rationale.

Tools

ToolStatusHeuristicDefault thresholds
check_ruminationlive (v0.0.1)word_overlap_jaccardwindow 90 min, count 3, similarity 0.55
check_hyperfocuslive (v0.0.2)elapsed_threshold_with_eod60 / 90 / 120 minutes
check_sycophancylive (v0.0.2)pattern overlap (4 patterns)similarity 0.5

Design invariants

  • Stateless. The server persists nothing — no SQLite, no JSONL, no in-memory caches that survive a call. Callers supply all history.
  • No telemetry, no network sockets. Per ETHICS.md commitment 4.
  • No user content in logs. Only tool_invoked metadata is logged.
  • Override-token vocabulary is closed at v0.1.0. New tokens require a minor bump and sign-off per ADR 0006 §3 and §10.
  • Heuristics are auditable. Source for each heuristic lives in src/neurodock_mcp_guardrail/heuristics/. Changes there require sign-off per ETHICS.md commitment 3.

Usage

bash
# Run the server over stdio (via the console script):neurodock-mcp-guardrail
# Or smoke-test the build directly:python -c "from neurodock_mcp_guardrail import server; print(server.app.name)"

Tests

bash
uv run pytest packages/mcp-guardrail/tests/ -v

Tests cover:

  • The Jaccard heuristic (identical, disjoint, stopword-only, paraphrase, case-insensitivity, punctuation handling).
  • The check_rumination tool (empty history, in-window detection, out-of-window non-detection, threshold honour, override-token contract, history-ordering rejection, false-positive-feedback path).
  • check_hyperfocus (escalation level mapping, end-of-day strictness, idle-signal handling, override-token contract, schema conformance).
  • check_sycophancy (each of the four patterns, counter-prompt generation, override-token contract, schema conformance).
  • Protocol conformance: every tool response is validated against the schema files in schemas/.

License

AGPL-3.0-or-later.

Source: packages/mcp-guardrail/README.md at commit 2d9cac6

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.0.5LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts