perf: parse the command line once for the command being run

[LadyBluenotes]

🎯 Changes

• cac finds the matched command by parsing the arguments one time for each registered command, and it does not stop at the match. Intent registers 15 commands, so every call parsed its arguments 15 times.
• main() now gives cac only the command named by the first argument for the parse, then restores the full command list. cac parses one time. When the first argument is a flag or an unknown command, nothing changes.
• The first main() call in a fresh process is 0.19 ms faster at the minimum (1.56 ms to 1.37 ms over 40 processes each). The load benchmarks, which call main() in a warm loop, are 7% to 9% faster. The list scan benchmark does not change.
• Help output, intent help <command>, unknown commands, and exit codes are the same.
• The workaround carries a marker that names the cac behavior and when to remove it. dist/cli.mjs grows by 181 bytes.
• Adds a patch changeset.

✅ Checklist

• I have followed the steps in the Contributing guide.
• I have tested this code locally with pnpm run test:pr (run pnpm build:all first).

🚀 Release Impact

• This change affects published code, and I have generated a changeset.
• This change is docs/CI/dev-only (no release).

Summary by CodeRabbit

• Performance
  • CLI argument parsing now processes only the command being run, reducing parsing overhead.

[changeset-bot]

🦋 Changeset detected

Latest commit: 71bf891

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 1 package

Name	Type
@tanstack/intent	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

[coderabbitai]

Navigate logical layers of code changes, visualize relationships, and explore their blast radius.

Note

Currently processing new changes in this PR. This may take a few minutes, please wait...

⚙️ Run configuration

• Configuration used: defaults
• Review profile: CHILL
• Plan: Advanced
• Run ID: 2a79e84a-0c17-47f6-b0f4-444cd97ae8cd

📥 Commits

Reviewing files that changed from the base of the PR and between db658f7 and c237754.

📒 Files selected for processing (2)

• .changeset/cli-single-command-parse.md
• packages/intent/src/cli.ts

 __________________________________
< Code Wars Episode IV: A New Bug. >
 ----------------------------------
  \
   \   \
        \ /\
        ( )
      .( o ).

✨ Finishing Touches

📝 Generate docstrings

• Commit to this branch
• Create a new PR

🧪 Generate unit tests (beta)

• Commit to this branch
• Create a new PR

• Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[nx-cloud]

View your CI Pipeline Execution ↗ for commit c237754

Command	Status	Duration	Result
nx affected --targets=test:eslint,test:sherif,t...	✅ Succeeded	1m 5s	View ↗
nx run-many --targets=build	✅ Succeeded	3s	View ↗

---

☁️ Nx Cloud last updated this comment at 2026-10-04 21:50:58 UTC

[pkg-pr-new]

Open in StackBlitz

npm i https://pkg.pr.new/@tanstack/intent@322

commit: 71bf891

[codspeed]

Merging this PR will improve performance by 19.83%

⚡ 4 improved benchmarks
✅ 8 untouched benchmarks

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
⚡	loads a direct dependency skill	33.1 ms	25.5 ms	+29.41%
⚡	loads a direct dependency from a large workspace	43.2 ms	36.1 ms	+19.88%
⚡	loads direct dependency content as json	46.2 ms	38.8 ms	+19.08%
⚡	checks a shipped skills tree	172.3 ms	154.4 ms	+11.6%

Tip

Curious why performance improved? Comment @codspeedbot explain why performance improved on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing perf/cli-single-command-parse (71bf891) with main (db658f7)¹}

Footnotes

1. No successful run was found on main (7a30838) during the generation of this report, so db658f7 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩