How do I install the Build skill?

Run npx skillmds@latest add tomevault-io/build in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Build safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Build?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Build free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Build by tomevault-io — Agent Skill

Name: Build
Author: tomevault-io

Build

Compile the TMDb Swift package for the current platform to check it builds. Use to verify code compiles after changes — delegates the build to a Haiku subagent and returns a concise pass/fail + errors-as-file:line summary, keeping logs out of context. To also compile the test targets, use /build-for-testing. Use when this capability is needed.

tomevault-io Updated Sep 17, 2026

File contents

Build project

Delegate the build to a Haiku subagent so the (potentially verbose) build output stays out of your context. Use the Agent tool with subagent_type: general-purpose and model: haiku, give it the prompt below, then relay its report. Do not run the build yourself.

Subagent prompt:

Build the TMDb Swift package and report the result concisely.

- If the xcode-tools MCP is available (running inside Xcode), run
  `mcp__xcode-tools__BuildProject`. On failure, get per-file error detail with
  `mcp__xcode-tools__XcodeRefreshCodeIssuesInFile` on each flagged file.
- Otherwise run `mkdir -p .build && make build > .build/last-build.log 2>&1`,
  check the exit status for pass/fail (the Makefile sets `pipefail`, so a compile
  failure propagates through the xcsift pipe), and summarise from that log file.
  **Trust the exit status, not xcsift's toon summary.** Outside a docs build,
  SwiftPM emits a benign `found N file(s) which are unhandled … *.docc` warning
  (the DocC plugin loads only under `SWIFTCI_DOCC=1`); xcsift lists it under
  `errors[]` with `null,null` coordinates and may print `status: failed`, yet the
  build still **exits 0**. A 0 exit whose only errors are these `.docc`
  unhandled-file entries is a **PASS** — report succeeded and exclude them from
  the error count.

Report back ONLY:
- Status: succeeded or failed
- Error and warning counts
- Each error/warning as `file:line — message` (omit this list if there are none)
- On failure, the full log path `.build/last-build.log` (or, inside Xcode, the
  flagged files so the caller can refresh their diagnostics)

Do not paste raw build logs or successful-compilation output.

If the report is unclear on a failure, read the log path it provides (or, inside Xcode, refresh diagnostics on the flagged files) rather than re-running. After fixing the issues, re-invoke this skill to re-check (a fresh subagent will rebuild).

Source: adamayoung/TMDb — distributed by TomeVault.

---name: builddescription: Compile the TMDb Swift package for the current platform to check it builds. Use to verify code compiles after changes — delegates the build to a Haiku subagent and returns a concise pass/fail + errors-as-file:line summary, keeping logs out of context. To also compile the test targets, use /build-for-testing. Use when this capability is needed.---# Build projectDelegate the build to a **Haiku subagent** so the (potentially verbose) buildoutput stays out of your context. Use the Agent tool with`subagent_type: general-purpose` and `model: haiku`, give it the prompt below,then relay its report. Do **not** run the build yourself.Subagent prompt:```textBuild the TMDb Swift package and report the result concisely.- If the xcode-tools MCP is available (running inside Xcode), run `mcp__xcode-tools__BuildProject`. On failure, get per-file error detail with `mcp__xcode-tools__XcodeRefreshCodeIssuesInFile` on each flagged file.- Otherwise run `mkdir -p .build && make build > .build/last-build.log 2>&1`, check the exit status for pass/fail (the Makefile sets `pipefail`, so a compile failure propagates through the xcsift pipe), and summarise from that log file. **Trust the exit status, not xcsift's toon summary.** Outside a docs build, SwiftPM emits a benign `found N file(s) which are unhandled … *.docc` warning (the DocC plugin loads only under `SWIFTCI_DOCC=1`); xcsift lists it under `errors[]` with `null,null` coordinates and may print `status: failed`, yet the build still **exits 0**. A 0 exit whose only errors are these `.docc` unhandled-file entries is a **PASS** — report succeeded and exclude them from the error count.Report back ONLY:- Status: succeeded or failed- Error and warning counts- Each error/warning as `file:line — message` (omit this list if there are none)- On failure, the full log path `.build/last-build.log` (or, inside Xcode, the flagged files so the caller can refresh their diagnostics)Do not paste raw build logs or successful-compilation output.```If the report is unclear on a failure, read the log path it provides (or, insideXcode, refresh diagnostics on the flagged files) rather than re-running. Afterfixing the issues, re-invoke this skill to re-check (a fresh subagent willrebuild).---> Source: [adamayoung/TMDb](https://github.com/adamayoung/TMDb) — distributed by [TomeVault](https://tomevault.io).

Frequently asked questions