# io.github.farssaadeldin7-ux/neural-link-intention-layer (mcpb · neural-link-intention-layer.mcpb)

Learn a designer's repetitive command patterns and surface the next likely action, cutting the...

- Trust score: 35/100 (low)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-10-24

## Components

- mcpb · `neural-link-intention-layer.mcpb`: 35/100 (this document), [markdown](https://verifymcp.io/servers/farssaadeldin7-ux-neural-link-intention-layer/neural-link-intention-layer.md), [page](https://verifymcp.io/servers/farssaadeldin7-ux-neural-link-intention-layer/neural-link-intention-layer)

## Channel facts

- Registry: `mcpb`
- Package: `https://github.com/farssaadeldin7-ux/claude-plugin-suite/releases/download/mcpb-v0.1.3/neural-link-intention-layer.mcpb`
- Version: `0.1.3`
- Transport: `stdio`
- SHA-256: `93db2f2cd060a8a52a42785c128bd1703e182986b55d5fb7cb05816797f8d845`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-10-24.

- **Supply Chain Security**: 0/100
  - No malware scan is available for this kind of package: the supply-chain vendors we use do not cover it. This is a permanent gap in our coverage, not a finding about the package.
  - Known CVEs could not be checked: we found no dependency list to check in this bundle: either it runs a compiled binary, whose bundled manifests we do not read, or none of the package.json, requirements.txt, pyproject.toml or uv.lock files it ships for its runtime declares a dependency we can read (a requirements.txt that only points at a setup.py project, a local path, or a nested requirements file we cannot follow counts as none).
  - Install-script risk not yet assessed.
  - Dependency health could not be checked: we found no dependency list to check in this bundle: either it runs a compiled binary, whose bundled manifests we do not read, or none of the package.json, requirements.txt, pyproject.toml or uv.lock files it ships for its runtime declares a dependency we can read (a requirements.txt that only points at a setup.py project, a local path, or a nested requirements file we cannot follow counts as none).
- **Provenance & Transparency**: 32/100
  - Source repository is publicly reachable at the declared URL.
  - Provenance check failed: no build-provenance attestation is published.
  - License check failed: no license is declared.
  - Actively maintained (last published 3 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 80/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 2043 tokens (~136/item across 15 items; 15 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability not yet verified: not enough scan history yet (needs a 30-day window).
- **Tool Coverage**: 98/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 93% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 15 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 16 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 20/100
  - Spec-recency check failed: implements MCP spec 2024-11-05; the latest is 2026-07-28.

**Unverified: 2 categories.** Categories scored 0 because we could not verify them: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.

## Install

### How do I install the io.github.farssaadeldin7-ux/neural-link-intention-layer MCP server?

io.github.farssaadeldin7-ux/neural-link-intention-layer ships as an MCPB bundle, a single file you download and open in an MCP host that supports MCPB bundles, such as Claude Desktop. The host reads the launch command from the bundle's own manifest, so there is no command to copy. Before opening it, check the file's SHA-256 matches the one the MCP registry declares for it, 93db2f2cd060a8a52a42785c128bd1703e182986b55d5fb7cb05816797f8d845; the commands to do that are on this page. If it does not match, do not open it.

- Download bundle: `https://github.com/farssaadeldin7-ux/claude-plugin-suite/releases/download/mcpb-v0.1.3/neural-link-intention-layer.mcpb`
- SHA-256: `93db2f2cd060a8a52a42785c128bd1703e182986b55d5fb7cb05816797f8d845`

### Verify on macOS / Linux

```bash
echo "93db2f2cd060a8a52a42785c128bd1703e182986b55d5fb7cb05816797f8d845  neural-link-intention-layer.mcpb" | shasum -a 256 -c
```

### Verify on Windows (PowerShell)

```powershell
(Get-FileHash -Algorithm SHA256 .\neural-link-intention-layer.mcpb).Hash -eq "93db2f2cd060a8a52a42785c128bd1703e182986b55d5fb7cb05816797f8d845"
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-10-21 (score 35)

First indexed and scored.

## MCP tools (15)

### `instrumentation_guide` (~114 tokens)

How to get a command history out of an application: the three tiers of source with their coverage and blind spots, per-application methods (Photoshop, Blender, Figma, After Effects, Illustrator), OS-level hotkey logging, the capture rules that keep a log private, and the exact log format the analysis tools expect. Reference data only — it measures nothing itself.

Input parameters:

- `application` (string): Optional: photoshop, blender, figma, after_effects, illustrator or other. Omit for the full guide.

### `automation_catalogue` (~101 tokens)

What each automation mechanism can and cannot do — Photoshop and Illustrator Actions, Figma components and the plugin API, Blender keymaps, After Effects expressions and scripting, OS-level macro tools, hardware — with setup costs and break-even frequencies. Reference data only — choosing between them is the skill's job.

Input parameters:

- `application` (string): Optional: photoshop, illustrator, figma, blender, after_effects, os_level or hardware. Omit to list everything.

### `method_reference` (~69 tokens)

The method's fixed numbers: normalisation rules, minimum log sizes, the evaluation protocol and expected accuracy ranges, the undo and navigation reading bands, the payback formula with its constants, and the confidence-floor table. Everything the licensed tools compute with, stated openly so the approach can be judged before buying.

### `analyse_log` (~175 tokens)

The sequence audit over a supplied command log: normalisation (runs collapsed, sequences cut at 3-minute idle gaps), top actions, top bigrams and trigrams, recurring sequences of length 4 or more, the undo diagnostic and the navigation share, with the reading bands attached. Counting only — it does not decide what to automate, and it refuses logs under 500 actions rather than reporting numbers they cannot support. Requires a paid plan.

Input parameters:

- `log` (string, required): One action per line: "timestamp application action", timestamp and application optional. Action names, counts and timestamps only — never document names, layer names or file paths.
- `navigation_actions` (array): Optional exact action names to count as navigation, on top of the default zoom/pan/navigate/visibility name patterns (tool toggles, for instance).

### `fit_predictor` (~271 tokens)

Fit the n-gram predictor on a supplied log and report its accuracy honestly: chronological split with the last 20% held out, baseline top-1, model top-1 with self-transitions excluded, top-3, and the log's date range and size — plus the contexts that clear the confidence floor (default 0.80), with destructive continuations suppressed. Requires at least 2,000 normalised actions; under 5,000 the trigram figures are provisional. With save: true, the fitted counts are also persisted locally so predict_next can answer later without refitting. It predicts the user's own recorded habits, not intention, and nothing is auto-executed. Requires a paid plan.

Input parameters:

- `k` (number): Optional cost asymmetry: how many times worse a wrong suggestion is than a right one is good. Floor is k / (1 + k). Default 4, giving 0.80.
- `log` (string, required): One action per line: "timestamp application action", timestamp and application optional.
- `save` (boolean): Optional: persist the fitted model (counts, fit metadata, holdout accuracy) to the local config directory, replacing any previously saved model, so predict_next and model_status can use it. The model…

### `predict_next` (~221 tokens)

The top-k next-action predictions from the saved model, given the most recent actions: each with its probability and whether it clears the confidence floor (default 0.80), destructive continuations listed separately and never surfaced, and the model's holdout accuracy attached so the caller knows how far to trust it. Same trigram-with-backoff arithmetic the holdout evaluation measured — it predicts the user's own recorded habits, not intention, and nothing is auto-executed. Needs a model saved by fit_predictor with save: true. Requires a paid plan.

Input parameters:

- `k` (number): Optional cost asymmetry: how many times worse a wrong suggestion is than a right one is good. Floor is k / (1 + k). Default 4, giving 0.80.
- `recent_actions` (array, required): The most recent actions, oldest first, as normalised action names from the same vocabulary as the fitted log. The last one or two are the prediction context.
- `top_k` (integer): How many predictions to return. Default 3.

### `model_status` (~64 tokens)

The saved predictor model's vital signs: fit date, log size and date range, vocabulary size, holdout accuracy, and a staleness verdict against the refit rule's eight weeks. Reads the local model file only; nothing is refitted. Requires a paid plan.

### `score_candidate` (~327 tokens)

The payback arithmetic for one automation candidate: value = F x (K + C) / (S + R) and payback_weeks = S x 1.3 / (F x (K + C)), against the 8-week build threshold. Arithmetic over supplied terms only — F is only meaningful if it was read from a log, and the result says so. Requires a paid plan.

Input parameters:

- `c_seconds` (number): Context-switch cost in seconds: 0 hand stays put, 1.5 mouse-to-menu, 4 dialogue, 15 leaving the app. Default 0.
- `correct_fire_p` (number): Optional probability the automation fires correctly, 0-1. Supply with wrong_fire_severity_seconds to include R.
- `f_per_week` (number, required): Occurrences per week, from the log.
- `k_seconds` (number, required): Seconds saved per occurrence: keystroke 0.3, menu trip 1.5, modal dialogue 4.
- `setup_seconds` (number, required): Setup seconds: ~180 keymap, ~300 action, ~900 component set, ~1200 OS macro, ~5400 plugin. Maintenance (30% annualised) is added by the formula.
- `stable_across_log` (boolean): Whether the sequence is stable across the whole log. Without it the verdict stays unconfirmed.
- `wrong_fire_severity_seconds` (number): Optional severity of a wrong fire: 2 if easily undone, 30 if silently wrong. If destructive, do not build — R is unbounded.

### `record_build` (~245 tokens)

Record an automation that was actually built — mechanism, sequence, predicted weekly firing rate, payback — in the local build log, so the two-week re-measure has something to check against. Given a build_id and observed_f_per_week instead, records the follow-up measurement on an existing build. Requires a paid plan; the licence check is the only thing this tool sends anywhere.

Input parameters:

- `application` (string)
- `build_id` (string): Existing build to record a follow-up on. Omit when recording a new build.
- `mechanism` (string): E.g. "Photoshop Action", "Blender keymap entry", "Stream Deck button".
- `notes` (string)
- `observed_f_per_week` (number): Observed firing rate at the re-measure, from a fresh log. Only with build_id.
- `payback_weeks` (number): The scored payback, from score_candidate.
- `predicted_f_per_week` (number): Predicted weekly firing rate, from the log.
- `sequence` (string): The action sequence automated, e.g. "new_layer -> fill_50_grey -> blend_mode:soft_light".

### `review_builds` (~59 tokens)

The record of past builds with predicted against observed firing rates. Counting only — it flags nothing and removes nothing; whether an underfiring macro comes out is the user's call. Requires a paid plan.

Input parameters:

- `limit` (number): Default 20.

### `license_status` (~52 tokens)

Check the Neural-Link Intention Layer licence on this machine: plan, included capabilities, remaining quota and seat usage. Call this when a paid tool reports a licence problem, or when the user asks what their plan includes.

### `license_activate` (~69 tokens)

Store and activate a Neural-Link Intention Layer licence key on this machine. Use when the user pastes a licence key, or after a purchase completes.

Input parameters:

- `license_key` (string, required): The licence key, e.g. PS-XXX-00000-00000-00000-0000

### `start_checkout` (~76 tokens)

Begin a Neural-Link Intention Layer purchase. Returns a Stripe Checkout link. Call list_plans first if the user has not chosen a plan.

Input parameters:

- `email` (string): Email address for the receipt and licence.
- `plan` (string, required): Plan id: "pro" or "team". Use list_plans to see the options.

### `list_plans` (~26 tokens)

List the available Neural-Link Intention Layer plans, prices and what each one includes.

### `billing_portal` (~32 tokens)

Open the Stripe billing portal to change plan, update the card, or cancel the Neural-Link Intention Layer subscription.

## Diagnostics

Captured diagnostic sections: Provenance. The full working is on the page: https://verifymcp.io/servers/farssaadeldin7-ux-neural-link-intention-layer/neural-link-intention-layer#diagnostics

## Score history

- 2026-10-24: 35
- 2026-10-23: 35
- 2026-10-22: 35
- 2026-10-21: 35

## Common questions

### What is the io.github.farssaadeldin7-ux/neural-link-intention-layer MCP server?

io.github.farssaadeldin7-ux/neural-link-intention-layer is an MCP server listed in the public MCP registry as io.github.farssaadeldin7-ux/neural-link-intention-layer. Learn a designer's repetitive command patterns and surface the next likely action, cutting the... This page covers its MCPB bundle (https://github.com/farssaadeldin7-ux/claude-plugin-suite/releases/download/mcpb-v0.1.3/neural-link-intention-layer.mcpb).

### Is the io.github.farssaadeldin7-ux/neural-link-intention-layer MCP server safe to use?

io.github.farssaadeldin7-ux/neural-link-intention-layer scores 35 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the io.github.farssaadeldin7-ux/neural-link-intention-layer MCP server expose?

io.github.farssaadeldin7-ux/neural-link-intention-layer exposes 15 tools: instrumentation_guide, automation_catalogue, method_reference, analyse_log, fit_predictor, and 10 more. Their descriptions and schemas cost roughly 1,901 tokens of context every time the server is loaded.

## Links

- Repository: https://github.com/farssaadeldin7-ux/claude-plugin-suite
- Changelog RSS feed: https://verifymcp.io/servers/farssaadeldin7-ux-neural-link-intention-layer/neural-link-intention-layer.xml
- Changelog JSON feed: https://verifymcp.io/servers/farssaadeldin7-ux-neural-link-intention-layer/neural-link-intention-layer.json
- HTML version of this page: https://verifymcp.io/servers/farssaadeldin7-ux-neural-link-intention-layer/neural-link-intention-layer