Seiro MCP
About
Seiro MCP is an MCP server and Skills that enables autonomous build workflows for visionOS (Swift) apps using Codex CLI / App.
Details
- Author
- karad
- Categories
- Developer Tools
Jump to
3. Optional: install and run bundled skill
seiro-mcp skill install --dry-run seiro-mcp skill install
- The bundled skill name uses theseiro-mcp-prefix to avoid collisions.
- The bundled skill canonical source is.agents/skills/seiro-mcp-visionos-build-operator/, includingSKILL.md,agents/openai.yaml, and icon assets under.agents/skills/seiro-mcp-visionos-build-operator/assets/.
- Useseiro-mcp skill remove seiro-mcp-visionos-build-operatorto roll back.
- skill removereturnsnot_foundwithout failing when the skill is already absent.
- Verify compatibility withseiro-mcp --versionbefore skill operations.
- For this release line,seiro-mcp skill installdefaults toseiro-mcp-visionos-build-operator. Passing that skill name explicitly is still supported.
- seiro-mcp skill installinstalls the bundled skill into the local Codex skills directory and does not install the Seiro MCP server binary or configure MCP settings.
- Useseiro-mcp --help,seiro-mcp skill --help, andseiro-mcp --versionfor self-check.
seiro-mcp --help seiro-mcp skill --help seiro-mcp skill install --help
Alternative GitHub install path for Codexskill-installer:
- Use Codexskill-installerwith these arguments when you want to install the skill directly from the public GitHub repository:
- --repo karad/seiro-mcp
- --path .agents/skills/seiro-mcp-visionos-build-operator
Add an entry like the following to Codex CLI config (~/.codex/config.toml) to call the visionOS tools:
[mcp_servers.seiro_mcp] command = "/Users/<your-username>/.cargo/bin/seiro-mcp"
- Codex CLI does not expand${HOME}, so use absolute paths and replace<your-username>.
- Preferseiro-mcp config mcpto print this snippet with the actual installed binary path.
- By default, Seiro MCP readsseiro-mcp.tomlfrom the project current directory.
- Restart Codex CLI and confirmmcp listshows the visionOS tools.
- The MCP client must spawn the server as a child process and perform the RMCP handshake over stdio. Runningcargo rundirectly without a client will fail immediately.
- Example with Inspector:
npx @modelcontextprotocol/inspector seiro-mcp
Lightweight visionOS Build MCP for AI Coding Agents
Seiro MCP is a lightweight MCP server focused on visionOS development workflows for AI coding agents. It enables Codex CLI and other MCP clients to safely build, validate, and inspect visionOS projects through a dedicated set of MCP tools.
The goal of Seiro MCP is not to expose every Xcode capability. Instead, it intentionally focuses on the developer workflows that AI coding agents perform most frequently during day-to-day development. Today it provides visionOS build tools together with bundled Codex skill guidance, and over time it will expand with additional developer-focused utilities for spatial computing while remaining focused on development rather than release management.
- Focused visionOS development workflows for AI coding agents
- Safe build automation through explicit project constraints
- AI-friendly MCP interface for build, diagnostics, and artifacts
- Codex Skill integration for preferred visionOS build operation
- Predictable local development workflow
- Written in Rust
General-purpose Xcode automation is powerful, but AI coding agents usually need a focused subset of development operations: validate the environment, build a project, inspect failures, and retrieve artifacts.
Seiro MCP keeps that interface intentionally small. Smaller MCP tool surfaces are easier for agents to reason about, easier for humans to review, and safer to run in local development environments. The project prioritizes predictable developer workflows over release automation, signing orchestration, or full Xcode replacement.
Lightweight does not mean fewer features. It means exposing the right capabilities for AI-assisted development while deliberately leaving release-oriented workflows, signing orchestration, and broad Xcode automation outside the project's scope.
cargo install seiro-mcp --locked seiro-mcp config mcp seiro-mcp config project
- Paste the output ofseiro-mcp config mcpinto Codex CLI config (~/.codex/config.toml).
- Runseiro-mcp config projectfrom the target project root.
- Editseiro-mcp.tomlto allow the target project paths and schemes.
- Restart the MCP client and use the visionOS tools.
You are now ready to ask Codex to build your visionOS project with Seiro MCP.
- Rust 1.91.1 (recommendrustup override set 1.91.1)
- Cargo (thecargocommand must be available)
- Codex CLI
- Any MCP client (e.g., official MCP CLI / Inspector)
- git,bash/zsh
If DevToolsSecurity is disabled, enable it first:
$ DevToolsSecurity -status Developer mode is currently disabled. $ sudo DevToolsSecurity -enable
If you already installed an earlier version, upgrade tov0.5.1with:
cargo install seiro-mcp --locked --force --version 0.5.1 seiro-mcp --version
Release history and upgrade notes are published inGitHub Releases.
To refresh bundled skill guidance after upgrading:
seiro-mcp skill remove seiro-mcp-visionos-build-operator seiro-mcp skill install
Paste the output into Codex config (~/.codex/config.toml):
[mcp_servers.seiro_mcp] command = "/Users/<user>/.cargo/bin/seiro-mcp"
Then create the project-local Seiro MCP config from the target project root:
[visionos] allowed_paths = [] allowed_schemes = [] xcode_path = "/Applications/Xcode.app/Contents/Developer"
MCP_CONFIG_PATHand--configremain available for non-default config locations. When neither is set, Seiro MCP readsseiro-mcp.tomlfrom the process current directory.
3. Optional: install and run bundled skill
seiro-mcp skill install --dry-run seiro-mcp skill install
- The bundled skill name uses theseiro-mcp-prefix to avoid collisions.
- The bundled skill canonical source is.agents/skills/seiro-mcp-visionos-build-operator/, includingSKILL.md,agents/openai.yaml, and icon assets under.agents/skills/seiro-mcp-visionos-build-operator/assets/.
- Useseiro-mcp skill remove seiro-mcp-visionos-build-operatorto roll back.
- skill removereturnsnot_foundwithout failing when the skill is already absent.
- Verify compatibility withseiro-mcp --versionbefore skill operations.
- For this release line,seiro-mcp skill installdefaults toseiro-mcp-visionos-build-operator. Passing that skill name explicitly is still supported.
- seiro-mcp skill installinstalls the bundled skill into the local Codex skills directory and does not install the Seiro MCP server binary or configure MCP settings.
- Useseiro-mcp --help,seiro-mcp skill --help, andseiro-mcp --versionfor self-check.
seiro-mcp --help seiro-mcp skill --help seiro-mcp skill install --help
Alternative GitHub install path for Codexskill-installer:
- Use Codexskill-installerwith these arguments when you want to install the skill directly from the public GitHub repository:
- --repo karad/seiro-mcp
- --path .agents/skills/seiro-mcp-visionos-build-operator
Add an entry like the following to Codex CLI config (~/.codex/config.toml) to call the visionOS tools:
[mcp_servers.seiro_mcp] command = "/Users/<your-username>/.cargo/bin/seiro-mcp"
- Codex CLI does not expand${HOME}, so use absolute paths and replace<your-username>.
- Preferseiro-mcp config mcpto print this snippet with the actual installed binary path.
- By default, Seiro MCP readsseiro-mcp.tomlfrom the project current directory.
- Restart Codex CLI and confirmmcp listshows the visionOS tools.
- The MCP client must spawn the server as a child process and perform the RMCP handshake over stdio. Runningcargo rundirectly without a client will fail immediately.
- Example with Inspector:
npx @modelcontextprotocol/inspector seiro-mcp
2. Validate sandbox policy before building
mcp call validate_sandbox_policy '{ "project_path": "/Users/<user>/codex/workspaces/vision-app", "required_sdks": ["visionOS", "visionOS Simulator"], "xcode_path": "/Applications/Xcode.app/Contents/Developer" }'
- Ifstatus: "ok", proceed tobuild_visionos_app.
- Ifstatus: "error"or an MCP error, fix based on the code:
- path_not_allowed: add the project parent directory tovisionos.allowed_paths.
- sdk_missing: first inspectdetails.diagnostics(probe_mode,effective_required_sdks,detected_sdks_raw,detected_sdks_normalized), then install visionOS SDK from Xcode > Settings > Platforms.
- devtools_security_disabled: runDevToolsSecurity -enable.
- xcode_unlicensed: runsudo xcodebuild -license.
- disk_insufficient: ensure 20GB+ free space for the build.
mcp call inspect_xcode_sdks '{ "required_sdks": ["visionOS", "visionOS Simulator"], "xcode_path": "/Applications/Xcode.app/Contents/Developer" }'
- This read-only tool returnsmissing_required_sdksand the same SDK probe context used for sandbox validation.
- Recommended troubleshooting order:validate_sandbox_policydiagnostics ->inspect_xcode_sdks(optional) -> retry validate/build.
mcp call inspect_xcode_schemes '{ "project_path": "/Users/<user>/codex/workspaces/VisionApp/VisionApp.xcodeproj", "xcode_path": "/Applications/Xcode.app/Contents/Developer" }'
- Use this whenproject_pathorschemeis unknown.
- Ifproject_pathis omitted, resolution order is:
- .xcodeprojdiscovered in current working directory
- visionos.default_project_pathinseiro-mcp.toml
3. Start a build withbuild_visionos_app
mcp call build_visionos_app '{ "project_path": "/Users/<user>/codex/workspaces/VisionApp/VisionApp.xcodeproj", "scheme": "VisionApp", "destination": "platform=visionOS Simulator,name=Apple Vision Pro", "configuration": "debug", "extra_args": ["-quiet"], "env_overrides": {"MOCK_XCODEBUILD_BEHAVIOR": "success"} }'
- project_path/workspacemust be absolute paths withinvisionos.allowed_paths.
- schememust be listed invisionos.allowed_schemes.
- configurationshould use lowercase canonical values:debugorrelease. For compatibility,Debug/Releaseare also accepted.
- Allowedextra_args:-quiet,-UseModernBuildSystem=YES,-skipPackagePluginValidation,-allowProvisioningUpdates.
- MOCK_XCODEBUILD_BEHAVIORswitches the test fixture (tests/fixtures/visionos/mock-xcodebuild.sh) amongsuccess/fail/timeout.
- On success, returnsjob_id,artifact_path,artifact_sha256,log_excerpt,duration_ms; on failure, returns errors such asbuild_failedortimeout.
- If multiple simulators match the destination name,build_visionos_appreturnsdestination_ambiguouswithmatched_devices,available_destinations, and a retry-readysuggested_destination.
If a build fails, inspect diagnostics without running manual shell commands:
mcp call inspect_build_diagnostics '{ "job_id": "<UUID returned in build error context>", "include_log_excerpt": true, "prefer_typecheck": true }'
- availability: "available"returnsprimary_location(file,line,column) from typecheck diagnostics.
- availability: "unavailable"falls back to anxcodebuild_logsummary with notes.
4. Download artifacts withfetch_build_output
mcp call fetch_build_output '{ "job_id": "<UUID returned by build_visionos_app>", "include_logs": true }'
- artifact_zippoints totarget/visionos-builds/<job_id>/artifact.zip; copy it beforedownload_ttl_secondsexpires.
- Setinclude_logs: falseto omitlog_excerptand reduce noise on the client side.
Seiro MCP keeps the MCP-only flow unchanged. You can choose either mode:
- MCP-only mode: callvalidate_sandbox_policy/build_visionos_app/inspect_build_diagnostics(on failure) /fetch_build_outputdirectly.
- Skill-assisted mode: use theseiro-mcp-visionos-build-operatorskill for Xcode / visionOS project workflows so Codex prefers Seiro MCP over directxcodebuild/swiftc.
- Ifproject_pathorschemeis missing in skill-assisted mode, runinspect_xcode_schemesfirst as optional preflight.
- When discovering a project locally, remember that.xcodeprojand.xcworkspaceare directory packages. Do not rely on file-only searches such asrg --filesto decide they are absent.
- .agents/skills/seiro-mcp-visionos-build-operator/SKILL.md
seiro-mcp skill install --dry-run seiro-mcp skill install
Install from GitHub with Codexskill-installer:
- --repo karad/seiro-mcp
- --path .agents/skills/seiro-mcp-visionos-build-operator
- Use seiro-mcp-visionos-build-operator for this visionOS build task.
- Please run this using the seiro-mcp-visionos-build-operator skill.
- Use Seiro MCP for this Xcode project instead of direct xcodebuild.
- Skills provide orchestration guidance.
- MCP provides execution capability.
- In skill-assisted mode, the actual execution remains MCP tool calls with unchanged contracts.
- Installing the skill from GitHub orseiro-mcp skill installdoes not install the Seiro MCP server binary or configure the MCP client connection.
- For Xcode / visionOS project tasks, direct shellxcodebuild/swiftcshould be treated as fallback paths, not the default path.
- The server must be launched as a child process by an MCP client; runningcargo rundirectly will fail withMCP_CLIENT_REQUIRED(exit 44).
- Seiro MCP currently supports local stdio MCP startup. TCP mode is not part of the supported local workflow.
- Seedocs/runbook.mdfor the full startup recipe.
- --config/MCP_CONFIG_PATH:--configwins; otherwiseMCP_CONFIG_PATH->./seiro-mcp.toml(relative paths are resolved to absolute).
- Token setup is not required for the default local Codex workflow.
- Exit codes:
- 44:MCP_CLIENT_REQUIRED(stdin/stdout is a TTY; must be launched via MCP client)
- Config file not found: runseiro-mcp config projectin the project root or set an absoluteMCP_CONFIG_PATH.
- MCP_CLIENT_REQUIRED: occurs when runningcargo rundirectly; always launch via an MCP client (Inspector / Codex, etc.).
- seiro-mcp: command not found: verify installation and useseiro-mcp config mcpto print the Codex MCP snippet.
- path_not_allowed: add the project parent tovisionos.allowed_pathsand restart.
- scheme_not_allowed: add the scheme tovisionos.allowed_schemesand restart.
- sdk_missing: checkdetails.diagnosticsfirst; ifprobe_modeisenv, verifyVISIONOS_SANDBOX_SDKS. Then runinspect_xcode_sdksand retry after SDK/config fixes.
- build_failed: usejob_idfrom the structured error and callinspect_build_diagnosticsto identify file/line before retrying.
- visionOS quickstart:docs/quickstart.md
- runbook:docs/runbook.md
- Configuration details:docs/config.md
As AI coding agents become more capable, they also need reliable ways to interact with local development environments. Existing solutions often aim to expose broad Xcode functionality, but many autonomous development tasks require only a small, well-defined subset of those capabilities.
Seiro MCP started from a simple idea: provide only the tools that are actually needed for AI-assisted visionOS development, and make those tools reliable, secure, and easy for AI agents to use. Rather than becoming a full-featured Xcode automation server, Seiro MCP is designed to be a focused development companion that helps AI agents build, validate, inspect diagnostics, and support spatial computing projects.
The long-term vision is to grow Seiro MCP into a collection of carefully designed developer tools that improve AI-assisted development for visionOS and spatial computing while preserving its lightweight philosophy.
Seiro MCP will continue to grow as a focused development toolkit for AI-assisted spatial computing work. Planned directions include:
- Additional visionOS developer tools
- Project analysis utilities
- Additional development workflows for visionOS
- Swift Package support
- Better AI agent workflow guidance
- More spatial computing developer utilities
The roadmap remains aligned with the lightweight philosophy: Seiro MCP should expose the right development tools for AI agents, not become a full-featured Xcode MCP server.
src/ lib/ # shared logic: errors, telemetry, filesystem helpers server/ # config + RMCP runtime tools/ # visionOS tools tests/ integration/ # integration tests (separate crate) docs/ # configuration, runbook, review checklists
If you are developing this repository itself, use the clone flow:
git clone git@github.com:karad/seiro-mcp.git cd seiro-mcp cargo fetch cargo run -p xtask -- langscan cargo run -p xtask -- docs-langscan cargo run -p xtask -- check-docs-links cargo run -p xtask -- preflight
- On success,target/release/seiro-mcpis produced.
cargo check cargo test --all -- --nocapture cargo fmt -- --check cargo clippy -- -D warnings cargo build --release cargo package --list cargo publish --dry-run
--lockedis recommended for reproducibility, but not mandatory for all environments.
- Preferred:cargo run -p xtask -- preflight(runs fetch/check/test/fmt/clippy/build in order).
- Manual:cargo fetch->cargo check->cargo test --all->cargo fmt -- --check->cargo clippy -- -D warnings->cargo build --release.
- Unit tests insrc/server/config/mod.rscover configuration validation (success and error cases).
- tests/integration/visionos_build.rscoversvalidate_sandbox_policy,build_visionos_app,inspect_build_diagnostics, andfetch_build_output, including TTL behavior.
- License:LICENSE
- Contributing:CONTRIBUTING.md
- Code of Conduct:CODE_OF_CONDUCT.md
- Security:SECURITY.md
This is a web browser that enables your coding agent, such as Claude Code, to visit websites on your behalf and assist you in identifying bugs or creating UI test cases.
An MCP server to help AI assistants to answer questions and generate AccelByte Extend SDK code more effectively .
Local stdio MCP server that lets AI coding agents read and maintain structured architecture, rules, and decisions directly from your repository.
Official Context7 MCP server that brings up-to-date, version-specific library documentation and code examples into AI coding prompts.
Remote, no-auth MCP server providing AI-powered codebase context and answers
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.





