agentleFS
Sign inStart free
lilly-buildslilly-builds/senior-engineer-mode/skills/lets-hack/SKILL.md

lets-hack

Skill0 starsChanged 5 months agoMITView on GitHub

Front door to the senior-engineer dev flow. Drives one code change through SCOPE → BUILD → VERIFY → REVIEW → SHIP, stopping at human gates and handing each step to the best existing tool. Use when starting a coding task or when the user wants to work "the spine." Say /lets-hack, or /lets-hack <verb> to jump to a step (scope | build | verify | review | ship).

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If the user names a verb (/lets-hack review), jump straight to that step.

    line 12
  2. Distinctive:

    Keep the diff small and focused.

    line 21
  3. Very specific:

    Run it through the actual route / button / endpoint / job — not just a helper.

    line 25
  4. Distinctive:

    When the UI looks wrong, check the data/response shape first.

    line 26
  5. Distinctive:

    Commit only after checks are green (the verify-before-done guard enforces this on a "done" claim).

    line 33

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Let's Hack (/lets-hack)

The front door to senior-engineer mode. Drive one code change through the five moves, in order. You don't reinvent tools here — each verb hands off to the best one that already exists. Stop at every human gate; do exactly what was asked, nothing more.

If the user names a verb (/lets-hack review), jump straight to that step. Otherwise start at SCOPE.

SCOPE — understand before editing

  • Enter plan mode. Send an Explore scout to find the real entry path and an existing pattern to reuse.
  • Decide the SMALLEST change; name the file(s) you'll touch and why. Batch any clarifying questions now.
  • Gate: present the plan (ExitPlanMode) and get sign-off before editing.

BUILD — the smallest correct change

  • Isolate risky or parallel work in a worktree (/worktree).
  • Keep the diff small and focused; match existing conventions; add nothing that wasn't asked.
  • The block-suppressions guard is live — fix real type/lint errors, don't reach for any / @ts-ignore.

VERIFY — prove it through the real path

  • Run it through the actual route / button / endpoint / job — not just a helper. Use /verify or /run.
  • When the UI looks wrong, check the data/response shape first. Show the command + output as evidence.

REVIEW — fresh eyes that didn't write it

  • Hand the diff to the skeptic subagent (or /code-review). Ask for correctness / requirement gaps only.
  • Triage: fix the real findings, skip the style nitpicks. Don't gold-plate.

SHIP — gate it, hand it off clean

  • Commit only after checks are green (the verify-before-done guard enforces this on a "done" claim).
  • Open the PR; clean up the branch/worktree as part of merging.
  • End with the honest report: what changed, why it's safe, what you verified, what's still risky, what you did NOT touch. Decisions about people, data, or money go to a human — don't auto-act.

Notes

  • The rules (~/.claude/senior-engineer-mode/rules/AGENTS.md) already load every session; this skill is the active walkthrough for when you want to drive the flow deliberately, step by step.
  • Full SOP: ~/.claude/senior-engineer-mode/THE-SPINE.md.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free