Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/oaslananka/kicad-mcp-pro/securitygit clone --depth 1 https://github.com/oaslananka/kicad-mcp-proWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/oaslananka/kicad-mcp-pro/security)<a href="https://agentmods.dev/agents/oaslananka/kicad-mcp-pro/security"><img src="https://agentmods.dev/badge/agents/oaslananka/kicad-mcp-pro/security.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00710 |
| Opus 5 | $0.00000 | $0.00355 |
| Sonnet 5 | $0.00000 | $0.00142 |
| Haiku 4.5 | $0.00000 | $0.00071 |
Grade A, and why
security scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Security Guide
Overview
KiCad MCP can read and modify PCB/schematic project files when write tools are enabled. This guide explains the security model and best practices.
Principles
- Default deny — Always start with read-only mode.
- Least privilege — Enable only the tools needed for the task.
- Explicit consent — Every destructive operation requires user approval.
- Defense in depth — Multiple layers of protection: operating mode, tool filtering, sandboxing, and path validation.
Operating Modes
| Mode | Description | When to Use |
|---|---|---|
readonly |
All write tools return denial | Default for untrusted projects |
write |
Write tools available with approval | Trusted projects, active design |
manufacturing |
Export tools + quality gates | Release preparation |
experimental |
All tools including auto-fix loops | Advanced users only |
Path Safety
KICAD_MCP_PROJECT_DIRis required and enforced- Path traversal attacks are blocked
- Workspace-relative paths are validated
- No write access outside the project directory
KICAD_MCP_OUTPUT_DIRlimits export destinations
Prompt Injection Defense
KiCad project files (schematic text, net names, component values, BOM notes) are untrusted inputs:
- No system prompt override from project file content
- No command injection from net names, reference designators, or file names
- DRC/ERC reports treated as data, not instructions
- BOM supplier URLs treated as untrusted
- All output paths sanitized before use
Remote Web Apps
Remote web apps (ChatGPT, Claude.ai) cannot directly access your local KiCad installation unless you explicitly run a local bridge:
- Remote tools are read-only by default
- Uploaded project archives are analyzed in an isolated environment
- Local bridge requires explicit pairing with user approval
- All bridge sessions are short-lived with rotating tokens
Sandboxing
VS Code and other MCP hosts support sandboxing:
- Filesystem: Restrict read/write to specific directories
- Network: Restrict allowed domains for outbound connections
- Deny access to sensitive paths (
~/.ssh,~/.gnupg, etc.)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 95 lines · 0 tokens per session scan A accd5166db5f
security is an agent published in the GitHub repository oaslananka/kicad-mcp-pro (64 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 710 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
index
This section centralizes the instructions and client configuration surfaces used by coding agents and MCP-capable clients that work with KiCad Studio Kit.
client-configs
KiCad MCP Pro supports local stdio and Streamable HTTP MCP transports. The canonical copyable examples live under examples/mcp-clients/.
codex-support
KiCad Studio Kit supports Codex through the external MCP client path.
bom-reviewer
You are a reviewer subagent. Your job is to independently verify a BOM artifact produced by the circuit-design skill. You have access to all KiCad MCP tools.
code-reviewer
Use this agent when a major code change to the MCP server itself has been completed and needs to be reviewed against the original plan and project coding standards.
schematic-plan-reviewer
You are a reviewer subagent. Your job is to independently verify a schematic placement and wiring plan artifact. You have access to all KiCad MCP tools.