Agent skill · Audited from source
Building Simulink Models
skill · matlab/simulink-agentic-toolkitNOASSERTION licenceAudited 14 Oct 2026 · source 9090332613b6
Is it safe? Yes — SAFE, grade B (89/100). We read this skill's source at commit 9090332613b6 and found no filesystem, network or shell access and no hard-coded secrets. What it does: Builds and edits Simulink, System Composer, Stateflow, and Simscape models. Use when modifying model structure, parameters, ports, connections, or Stateflow chart internals.
Cite
Cite this page
ModelsAtlas. “Building Simulink Models (matlab/simulink-agentic-toolkit)”. Retrieved 14 Oct 2026. https://modelsatlas.com/skills/matlab/building-simulink-models
Trust audit
Audit 0.4.1 · source read 14 Oct 2026
SAFE means nothing in the source contradicts what the skill says it does. Grade B is shared by 13,985 of 14,816 audited skills.
What the source does
- FilesystemNone observedNo reads or writes in the source
- NetworkNone observedNo outbound calls in the source
- ShellNone observedNo commands run in the source
- DependenciesPinnedNothing floats
- SecretsNone foundNo keys in source
Five audit layers
| Layer | What we check | Result | Findings |
|---|---|---|---|
L0 Provenance | Repo owner, licence, archive status, activity | WARN | 1 |
L1 Static source | Secrets, path traversal, supply chain, hidden code | NA | 0 |
L2 Instructions | Prompt injection, authority framing, hidden directives | PASS | 1 |
L3 Inventory | The files and declarations match the source | PASS | 0 |
L4 Sandbox | Run in isolation and watch behaviour | SKIPPED why | 0 |
Gates applied: no behavioural pass.
Findings (2)
MEDIUMInventory / provenanceinv.binary · tools/common/+sage/+internal/+utils/@License/License.p
License.p
- Why
- a compiled or binary member cannot be reviewed from source
- Fix
- ship source, or explain the binary in the README
LOWObfuscation / stealthreview.concealment · SKILL.md
Do not narrate the gate check to the user; proceed silently based on the result.
- Why
- Directs the agent to withhold information about a prerequisite check from the user.
- Fix
- rewrite it so the instruction says plainly what it does, and asks the user before it acts
Audit 0.4.1 · source sha 9090332613b6 · 14 Oct 2026Report an issue or request a re-scan
Provenance
Who published it, and what the verdict is pinned to
- Publishermatlab GitHub organizationOwner of the repository this skill is read from
- Source foldermatlab/simulink-agentic-toolkit/skills-catalog/model-based-design-core/building-simulink-modelsThe exact folder we audited
- LicenceNOASSERTIONAs the repository states it
- Commit pinned for this verdict9090332613b6A new audit runs when the source changes
- Repository statusNot archivedLast push 6 days ago
- Sandboxed runNot runSkills are read, not executed; layer L4 is skipped
What it tells the agent
Full source on GitHub ↗Its description: Builds and edits Simulink, System Composer, Stateflow, and Simscape models. Use when modifying model structure, parameters, ports, connections, or Stateflow chart internals.
---
name: building-simulink-models
description: Builds and edits Simulink, System Composer, Stateflow, and Simscape models. Use when modifying model structure, parameters, ports, connections, or Stateflow chart internals.
license: https://www.mathworks.com/content/dam/mathworks/license/pmrl/license.md
metadata:
author: MathWorks
version: "1.5"
---
# Building Models
Use `model_edit` for all structural changes — Simulink, System Composer, Simscape, and Stateflow chart internals
## When to Use
- Adding, connecting, deleting, or replacing blocks in a model
- Configuring block parameters, signal properties, or model settings
- Creating or editing Stateflow chart internals (states, transitions, junctions, data, events, messages, functions)
- Building System Composer architecture models
- Wiring Simscape physical connections
## When NOT to Use
- Querying parameter values → use `model_query_params`
- Resolving variable references to numeric values → use `model_resolve_params`
## Library & Policy Prerequisites — BLOCKING GATE
**Do this FIRST — evaluate the result before any other tool call.** Do not narrate the gate check to the user; proceed silently based on the result.
Call `library.settingsLookup()` to check for custom library configuration. It returns a struct with resolved absolute paths:
- `found` — whether a libraries file was located (satk-libraries.json or legacy reuse-libraries.json)
- `enabled` — whether custom libraries are configured
- `gatePass` — whether the gate is satisfied (can proceed to build)
- `kgIndexPath` — absolute path to `library-kg/index.md`
- `librariesPath` — absolute path to the libraries file
- `policyPath` — absolute path to `block-policy.json`
- `dataRoot` — absolute path to the directory containing `.satk/`
**Interpret the result:**
- **`found: false`** → no custom libraries configured. Proceed with standard Simulink blocks.
- **`gatePass: true, found: true`** → read `kgIndexPath` for block lookup, then plan and edit.
- **`gatePass: false`** → invoke `setup-custom-libraries` to populate the Knowledge Index.
Display meaningful description of each check and result instead of stating gate Pass/fail.
No model reading, planning, or editing begins until the gate passes.
## Workflow
0. **Ensure Library & Policy Prerequisites:** See BLOCKING GATE above. Do not proceed until gates pass.
1. **Library block lookup:** List every block type you plan to use, search `.satk/library-kg/index.md` and the relevant category pages to find each of the library blocks that match.
2. **Read first:** Use `model_read` on the target scope to get block IDs and understand existing topology.
3. **Plan the data flow:** For complex edits, sketch inputs → operations → outputs, then map to blocks identified in Step 1 & 2.
4. **Edit:** Use `model_edit` with operations scoped to one subsystem level at a time.
5. **Verify:** Use `model_read` on the scope to confirm the structure matches your intent.
6. **Check connectivity:** After all edits in a scope are complete, run `model_check` to catch unconnected ports or dangling lines. Fix any `error`-severity issues.
**If `model_edit` returns `status: partial`:** Run both `model_read` and `model_check` immediately — don't wait until all edits are complete.
## Operation Chaining with `ref`
Use `ref` to name a block and `#ref` to reference it in later operations within the same call:
```json
[{"op": "add_block", "type": "Gain", "name": "MyGain", "ref": "g1"},
{"op": "connect", "target": "blk_5.y1 -> #g1.u1"}]
```
In SF scope, `#ref` references are portless — no `.y1`/`.u1` suffixes (see `references/stateflow.md`).
The response `created` map shows `ref → blk_id` (or `ref → sf_X` in SF scope). In subsequent calls, use the returned ID — `#ref` only works within a single call.
## Guardrails
- **Never manually construct block path strings** from names shown in `model_overview` or `model_read`. Block names can contain invisible newlines and trailing whitespace that cause `hilite_system`, `open_system`, and `get_param` to fail. Instead, resolve paths from `blk_X` IDs:
```matlab
% blk_42 → use the number after "blk_" as the SID
blockPath = Simulink.ID.getFullName('<ModelName>:42');
hilite_system(blockPath)
open_system(blockPath)
get_param(blockPath, 'BlockType')
```
- Do not call `Simulink.BlockDiagram.arrangeSystem` or use `set_param` for block positioning unless the user explicitly requests it. `model_edit` has a built-in autolayout engine that runs automatically after each call.
- Always pass `layout_mode` to `model_edit`. Use `"full"` when populating an empty scope (new model root, or a newly-created subsystem) for optimal block arrangement. Use `"incremental"` when adding blocks to a scope that already has existing blocks (preserves existing positions).
- Each `model_edit` call operates in exactly one domain (Simulink, System Composer, or Stateflow) determined by the scope. To add a Chart block and then populate it, use two calls: SL scope for the Chart block, then SF scope for internals. Use `model_read` between calls to discover the chart's `sf_X` scope ID.
- Use meaningfully named variables (e.g., `Kp_SpeedController`) instead of hardcoded numeric values. Define variables in model workspace or a `.m` init script.
- Don't use `evaluate_matlab_code` with `set_param`/`add_block` to bypass `model_edit` — it skips autolayout, undo tracking, and error recovery
- Use `open_system` rather than `load_system` to open models that are not already open, or when creating new models, unless the user explicitly asks otherwise or the model is a library. This ensures the user can see live edits as they happen.
## Naming Conventions
Prefer code-generation-safe names for blocks, signals, and variables:
- Use only: `a-z`, `A-Z`, `0-9`, underscore (`_`)
- Don't start with a number
- Don't use leading/trailing or consecutive underscores
- Prefer names under 32 characters (required for some code generation targets)
## Block Types
Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.
Install
Pinned to the audited commit 9090332
git clone https://github.com/matlab/simulink-agentic-toolkit git -C simulink-agentic-toolkit checkout 9090332613b6
The author documents no install command, so this uses the repository directly. Our verdict covers commit 9090332613b6 only.
Works in
The author has not said which agents this skill supports, and we have not tested it in any.
Audit history
Every audit, every source change
- ✓SAFE B 89 / 100
first audit · source
9090332613b6
README badge
For maintainers · shows the latest audit
[](https://modelsatlas.com/skills/matlab/building-simulink-models/)
The badge shows our verdict and grade for this skill, and links back to this page. It changes when the audit does.
Similar skills
matlab/simulink-agentic-toolkit
Generate a standalone bug report that another developer can use to reproduce, investigate, and fix an issue. Use when the user says 'file a bug', 'write a bug report', 'report this issue', or asks to document a defect…
matlab/simulink-agentic-toolkit
Generates draft requirements from Simulink models. Use when drafting or updating requirement artifacts from a model. Prefers Requirements Toolbox (.slreqx) when available; falls back to structured YAML.
matlab/simulink-agentic-toolkit
Configures Simulink simulations non-destructively using SimulationInput objects — parameter overrides without modifying the model, batch sweeps via parsim, custom input signals via Dataset, and simulation data retrieval…
matlab/simulink-agentic-toolkit
Specify algorithms for Model-Based Design: system specs, architecture specs, implementation plans, test plans. Use when creating specifications for controllers, signal processing, diagnostics, estimators, or other…
matlab/simulink-agentic-toolkit
Specify plant models for closed-loop simulation: system specs, architecture, build plans, validation plans. Use when creating, updating, or reviewing plant model specifications, planning plant model architecture, or…
matlab/simulink-agentic-toolkit
The Simulink Agentic Toolkit gives your AI agent both the tools and the expertise to work effectively with Simulink and Model-Based Design.
Add this badge to your README
[](https://modelsatlas.com/skills/matlab/building-simulink-models/)
<a href="https://modelsatlas.com/skills/matlab/building-simulink-models/"><img src="https://modelsatlas.com/skills/matlab/building-simulink-models/badge.svg" alt="ModelsAtlas: SAFE B"></a>
Questions about Building Simulink Models
What does the Building Simulink Models skill do?
Is Building Simulink Models safe to install?
What can Building Simulink Models access on my machine?
How current is this page?
Sources
- github.com/matlab/simulink-agentic-toolkit/tree/main/skills-cata — the audited source at commit 9090332613b6 and every finding aboveObserved · 14 Oct 2026
Spotted something wrong? Report a correction · Methodology
Keep going in your AI
Opens with this page as the source. We add nothing else to the prompt.
- What does the Building Simulink Models skill do?
- Is Building Simulink Models safe to install?
- What can Building Simulink Models access on my machine?