Grok build
Orchestrate coding work by delegating well-specified implementation tasks to xAI's Grok Build CLI (grok) running headlessly, while the coding assistant plans, writes the task specs, reviews every diff, and owns the result. Use when user says: 'use grok', 'grok build', 'delegate to grok', 'have grok implement', 'have grok execute', 'have grok build', 'send to grok', 'execute this plan with grok'. Executes a Markdown implementation plan task-by-task, or ad-hoc tasks with an inline spec.From its SKILL.md
npx -y skills add sanjay3290/ai-skills --skill grok-buildAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- skips confirmationTells the agent to proceed without asking first, 2 times: "run with --always-approve" and 1 more.
- runs commandsInstructs the agent to run 8 commands, including `grok update --check --json` and 7 more.
What its file declares
Copied from the file, not written here
The file declares its own license as Apache-2.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
5.4 KB, ~1.2k tokens by cl100k_base, as published. Nobody here has run it
Grok Build Orchestration
The coding assistant is the orchestrator: it plans, writes self-contained task specs,
dispatches them to Grok Build headlessly, reviews every diff, and owns the final result.
Grok is the fast, cheap executor. Full CLI details and verified behaviors: references/cli.md.
When to delegate vs keep with the orchestrator
| Delegate to Grok | Keep with the orchestrator |
|---|---|
| Plan tasks with clear acceptance criteria | Ambiguous requirements, architecture decisions |
| Boilerplate, scaffolding, CRUD | Deep cross-file debugging |
| Mechanical refactors | Security-sensitive code |
| Test writing from clear specs | Anything touching production infrastructure |
| UI components from mockups/specs | Tasks where writing the spec ≈ doing the work |
When in doubt, keep it with the orchestrator.
Session preflight (once, before the first dispatch)
grok update --check --json— ifupdateAvailableis true, rungrok updateand confirm withgrok --version.grok models— if it errors or reports logged out, STOP and ask the user to rungrok login.
Per-task loop (sequential — the default)
-
Spec. Write a self-contained task file (template below) to a temp directory OUTSIDE the target repo — the harness scratchpad if one is available, else the OS temp dir. Never write it inside the target repo. Grok has zero conversation context: no one-liner prompts, ever.
- POSIX:
mkdir -p "${TMPDIR:-/tmp}/grok-specs", then writetask.mdthere. - Windows (PowerShell):
New-Item -ItemType Directory -Force "$env:TEMP\grok-specs", then writetask.mdthere.
- POSIX:
-
Clean state. No uncommitted source changes — commit or stash first, so the post-run diff is exactly Grok's work. Ignore build artifacts (
__pycache__,dist/, etc.); if they show ingit status, they're usually just un-gitignored, not your concern. Never dispatch on a dirty source tree. -
Dispatch.
POSIX:
grok --prompt-file <task-file> \ --output-format json \ --always-approve \ --max-turns 30 \ --cwd <repo>Windows (PowerShell) — backtick line-continuation:
grok --prompt-file <task-file> ` --output-format json ` --always-approve ` --max-turns 30 ` --cwd <repo>Parse the JSON output and save
sessionId. (--always-approveis required for headless runs —--permission-mode acceptEditssilently cancels edits with no interactive approver. Seereferences/cli.md.) For a high-stakes task, add--checkso Grok self-verifies before you review; skip it otherwise (it ~doubles latency). -
Review gate — non-negotiable.
- Read the diff yourself (
git diff -- <files from the spec>to skip artifact noise): does it do the task, only the task, and match repo conventions? - Run the acceptance commands from the spec.
- Pass → commit with a clear message following the repo's convention → next task.
- Fail → fix-up:
grok --resume <sessionId> -p "<specific feedback>" --always-approve --output-format json. Max 2 fix-up rounds. Still failing → revert Grok's changes (git checkout -- .;git clean -fdfor new files), do the task yourself, and tell the user Grok couldn't complete it.
- Read the diff yourself (
Task spec template
# Task: <one-line title>
## Context
- Repo: <path> — <one line on what the project is>
- Conventions: <test runner, formatter, a good example file to imitate>
## Files
- Modify: <path>
- Create: <path>
## Task
<precise description of the change>
## Constraints
- Do not modify any files other than those listed above.
- <other constraints>
## Acceptance criteria
- `<exact command>` <expected result>
Executing a Markdown implementation plan
- One plan task per dispatch, in order.
- Check off the plan's task checkboxes (
- [ ]→- [x]) as each task lands and passes the review gate. - If the plan explicitly marks tasks as independent, see Parallel dispatch below; otherwise stay sequential.
Parallel dispatch (opt-in exception, not the default)
Only when a plan explicitly marks tasks independent: dispatch each with
--worktree=<task-slug>, run concurrently, then review and merge one worktree at a
time through the same review gate. Merge conflicts usually eat the savings — prefer
sequential.
Failure handling
| Failure | Action |
|---|---|
stopReason: "Cancelled", empty text, no diff | Missing --always-approve — retry with it |
| CLI error / timeout | Retry once; then do the task yourself and note the fallback |
| Auth expired | Stop; ask the user to run grok login |
| 2 fix-up rounds exhausted | Revert Grok's diff; the orchestrator finishes the task |
| Dirty tree at dispatch | Refuse; commit/stash first |
Models
Default grok-4.5. Add -m grok-composer-2.5-fast only for trivial mechanical tasks.
What ships with it: 1 file
4.1 KB alongside SKILL.md
references/
- cli.md4.1 KB
Gives 0 of the 12 instructions most plan spec skills give in ~1.2k tokens
Counted across 1,360 of the 2,617 authors here whose files we hold, read 2026-09-06
- Ask one question at a timein 73 of 1360
- Write the spec using the templatein 22 of 1360
- Ask clarifying questions if neededin 19 of 1360, across 18 files
- Wait for user confirmation before proceedingin 19 of 1360
- Save plans to the plans directoryin 17 of 1360, across 13 files
- Check for product marketing context firstin 16 of 1360, across 5 files
- Read the plan file completelyin 16 of 1360
- Order tasks by dependencyin 16 of 1360
- Gather context from the conversationin 15 of 1360, across 9 files
- Explore the codebase instead of askingin 15 of 1360, across 13 files
- Wait for explicit user approvalin 14 of 1360, across 13 files
- Quiz the user on the breakdownin 13 of 1360, across 7 files
Said here and by no other author read
- Write self-contained task specs to an external temp directory
- Commit or stash uncommitted changes before dispatch
- Run grok with the task prompt file and required flags
- Review the git diff and run acceptance commands
- Commit changes on pass or run fix-up rounds on fail
- Revert changes if fix-up rounds are exhausted
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.