Official agent skill

Implementation Kickoff

by openai in openai/openai-agents-python

Carry implementation through an isolated worktree and local handoff.

OfficialMITAuto-check passedDevelopment

Install Implementation Kickoff

skills CLI
$ npx skills add openai/openai-agents-python --skill implementation-kickoff -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install openai/openai-agents-python implementation-kickoff --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/openai/openai-agents-python.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/implementation-kickoff .claude/skills/implementation-kickoff && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
implementation-kickoff
GitHub stars
30k
Token cost
~2.9k tokens
SKILL.md length
1,608 words
Files
4 (incl. scripts)
Skills in repo
14
Repo updated
First seen
Licence
MIT

At a glance

Carry implementation through an isolated worktree and local handoff.

  • Works in 8 steps: Establish the task boundary → Create a detached worktree from current… → Implement without task commits → …
  • Tasks that involve Git worktrees
  • SKILL.md covers Non-negotiable boundaries, 1. Establish the task boundary, 2. Create a detached worktree… and 3. Implement without task…, plus 6 more sections
  • Runs Python scripts from its folder; calls git and python

What it does

Implementation Kickoff is an agent skill from openai/openai-agents-python, published by the product's own GitHub organization. Carry implementation through an isolated worktree and local handoff. Use only when this skill is explicitly invoked.

Its SKILL.md is about 2.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 5 other files, including scripts (for example `agents/openai.yaml`, `scripts/test_validate_handoff.py` and `scripts/validate_handoff.py`).

It sits in Development, covering Git worktrees. It works with OpenAI. The repository describes itself as: A lightweight, powerful framework for multi-agent workflows. The licence is MIT.

When your agent uses it

  • Tasks that involve Git worktrees

Example prompts

  • “/implementation-kickoff”

Requirements

  • Python 3

Workflow steps

8 steps, taken from the step headings in SKILL.md.

  1. Establish the task boundary
  2. Create a detached worktree from current main
  3. Implement without task commits
  4. Replay the complete task onto the latest main
  5. Complete final review and verification
  6. Generate the complete PR handoff
  7. Recheck main and create one commit
  8. Validate and hand off

What it can do on your machine

Read from SKILL.md and the folder at commit 71c2da4. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 2 files in scripts/ (Python), which the agent can run.

    Shell commands in SKILL.md call:

    • git
    • python

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Implementation Kickoff loads about 2.9k tokens when it runs. Until then it costs about 35 tokens; SKILL.md has 1,608 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~35
When it runs · the whole SKILL.md, loaded when a task matches
~2.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from openai/openai-agents-python at commit 71c2da4, republished under its MIT licence (© openai). 1,608 words, ~2,910 tokens.

Download SKILL.mdSave it as .claude/skills/implementation-kickoff/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
implementation-kickoff
description
Carry implementation through an isolated worktree and local handoff. Use only when this skill is explicitly invoked.

Implementation Kickoff

Use this skill as the explicit transition from an agreed implementation scope to isolated execution. Keep the user's original checkout and existing branches unchanged, and finish with a clean local branch that is ready for the user to push.

Non-negotiable boundaries

  • Treat explicit invocation of this skill as authorization to fetch, create one dedicated worktree, rebase or replay task-owned changes, create the final local branch, stage task-owned files, and create one local commit. It never authorizes push, pull-request creation, or any GitHub mutation.
  • Do not start during an investigation-only phase or before a required user approval. Finish planning and any required implementation scope contract first.
  • Use read-only GitHub access when remote PR evidence is required.
  • Preserve unrelated and user-owned changes. Do not remove an existing worktree or rewrite an existing branch to make room for this workflow.

1. Establish the task boundary

Record the original requirement, success criteria, intended target (origin/main unless the user states otherwise), task-owned paths, compatibility boundary, intentionally unsupported cases, and required repository skills. For a multi-step task, create and maintain the repository's required ExecPlan. An ExecPlan, review packet, ledger, trace, or temporary report is operational-only by default even when repository policy requires creating it; do not add it to the shipped-path manifest unless the original requirement or repository policy explicitly makes that exact path a committed deliverable.

If the current directory is a worktree previously created for this same task in the current conversation, resume it. Otherwise, continue from the user's current checkout only long enough to create a new worktree.

2. Create a detached worktree from current main

  1. Verify the source checkout's raw status without modifying it.
  2. Fetch origin main. If the fetch fails, stop rather than claiming a stale ref is current.
  3. Record the fetched origin/main commit.
  4. Choose a unique task-oriented path under the configured Codex worktree root. Check both the filesystem and git worktree list; never reuse or delete a collision.
  5. Run git worktree add --detach <worktree> origin/main and perform all subsequent implementation work there.
  6. Confirm the new worktree is detached at the recorded commit and initially clean.

Do not create the final branch yet. A detached worktree makes the eventual $pr-draft-summary branch suggestion authoritative and prevents temporary naming from becoming accidental output.

3. Implement without task commits

Keep the task diff uncommitted through implementation, focused tests, formatting, and review fixes. Track new files explicitly because ordinary diff statistics omit untracked files. Maintain one canonical shipped-path manifest separately from operational artifacts and require a concrete deliverable reason for every path in it. Use the applicable repository skills and references, including $implementation-strategy before user-facing or runtime changes.

When the task can be decomposed without temporarily breaking a supported contract, implement one narrow end-to-end behavior slice at a time and run its focused test before adding the next slice. Do not force cross-cutting migrations or atomic compatibility changes into artificial slices that cannot remain valid independently.

Do not create checkpoint commits. If an external interruption requires extra protection, leave the dedicated worktree intact or use a clearly named temporary stash; restore the changes before continuing and do not treat the stash as a deliverable.

Taking over an existing pull request

When the user asks to complete another author's pull request:

  1. Refresh the PR metadata, head, discussion, and complete three-dot diff through read-only access.
  2. Confirm that the PR is still an appropriate takeover source. Do not treat an already merged PR as an active takeover.
  3. Apply the original PR's complete task diff onto the worktree based on current origin/main; do not derive the final branch from the contributor branch and do not preserve its intermediate commit topology.
  4. Record the original PR number, PR author login, verified commit identity, existing valid Co-authored-by trailers, linked issues, and the original intent that the replacement must preserve.
  5. If a valid author identity cannot be obtained from the PR's commits, stop before committing and ask the user. Never invent an email address.

4. Replay the complete task onto the latest main

After implementation, focused tests, and formatting are stable, fetch origin main again. If it advanced:

  1. Confirm that every local change is task-owned.
  2. Save tracked and untracked task changes in a uniquely named temporary stash.
  3. Rebase the detached HEAD onto origin/main. With no task commits, this updates the empty local commit range to the new base.
  4. Reapply the stash and confirm it was removed only after a clean application.
  5. Resolve conflicts only when the requirement and surrounding source make the correct result unambiguous. Otherwise preserve the stash and conflict evidence, then stop for user direction.
  6. Rerun formatting and every focused check affected by the new base.

Record this observed origin/main commit as the final-base candidate. Do not call an older base "latest" merely because its changes appear unrelated.

5. Complete final review and verification

Run applicable completion gates against the complete task-owned diff on the final-base candidate. Apply formatting and safe hook-equivalent normalization before review, including generated-file provenance checks where relevant. Use $implementation-final-review to select lightweight, ordinary, or high-risk review; supply all task-owned untracked file contents as well as tracked changes. Only high-risk review requires review_state.py --complete-diff-output <complete.diff> and the strict packet/ledger protocol. Follow the selected procedure's evidence and invalidation rules.

Run $code-change-verification after clean review whenever its SDK eligibility rules apply. A lightweight review exemption does not waive an otherwise required SDK verification stack. Repo-meta work uses its applicable skill and focused checks. Do not create the branch or commit while required review, verification, or evidence remains incomplete.

Show full SKILL.md (681 more words)Show less

6. Generate the complete PR handoff

Invoke $pr-draft-summary only after review and verification apply to the final content. Give it a self-contained packet containing the original requirement, implementation scope contract, important decisions and intent, complete changed-path inventory including untracked files, final diff and statistics, compatibility notes, issue references, and takeover provenance.

The worktree is intentionally detached. Tell $pr-draft-summary to treat the current branch value HEAD as "no branch yet" and require a concrete unused branch-name suggestion; never accept HEAD as the suggestion. The description must explain the complete final change and its motivation, not only the last review fix.

For a takeover, begin the description with prose such as This pull request supersedes #<number> and .... Preserve any separate issue-closing line only when the final implementation actually resolves that issue.

If the diff, scope, base, behavior claim, issue relationship, or provenance changes after generation, regenerate the entire PR handoff.

7. Recheck main and create one commit

Fetch origin main once more immediately before creating the branch. If it differs from the final-base candidate, return to section 4 and replay onto the new base. Apply the selected review tier's base-change rules in $implementation-final-review; only high-risk work uses the verified base-advance closure in high-risk-review.md step 20. Recheck relevant upstream dependencies and tooling before retaining review evidence. Rerun applicable final verification on the new base and regenerate the PR handoff; obtain affected independent re-review whenever the selected tier requires it. Once stable:

  1. Check whether the suggested branch exists locally, remotely, or in another worktree. Ask $pr-draft-summary for the next available numeric suffix and regenerate the handoff before creating a colliding branch.
  2. Create the exact suggested branch in the task worktree.
  3. Stage only the task-owned shipped-path manifest, including intended new files. Compare the staged changed-path set byte-for-byte with the canonical shipped manifest before committing; any missing or unexpected path is a hard stop. In particular, do not stage an ignored ExecPlan or review artifact merely because it was required during implementation.
  4. Use the PR draft title as the commit subject.
  5. For a takeover, add the verified original PR author as Co-authored-by: Name <email>, retain distinct valid co-author trailers from the imported commits, and deduplicate identities.
  6. Create exactly one commit. Let repository hooks run normally.

Branch creation and committing identical content are repository bookkeeping and do not invalidate clean content review. If a hook or manual fix changes task content, stop, classify the change under $implementation-final-review, and rerun every invalidated gate and $pr-draft-summary before replacing or amending the commit.

8. Validate and hand off

Run python .agents/skills/implementation-kickoff/scripts/validate_handoff.py --repo <worktree> --base <final-base> --expected-branch <branch> --shipped-path-manifest <manifest>. For a takeover, also pass --required-trailer-email <verified-email> for each identity that must be credited. The shipped-path manifest must be a finite regular file whose type and content are read from one opened descriptor. It contains one exact repository-relative shipped path per line and excludes operational artifacts.

Independently confirm that the committed diff has the reviewed content fingerprint when final review supplied one. The validator checks Git topology and repository cleanliness; it does not replace semantic review or fingerprint verification.

Leave the worktree in place. Report the worktree path, final observed base commit, branch, commit SHA and subject, verification results, review status, PR title and description, and whether takeover provenance was included. Treat the worktree path and validator output as local diagnostics: never include them in the PR title, PR description, or other copy-ready external text. State explicitly that nothing was pushed and no pull request was created.

Failure behavior

  • Fetch failure: stop without creating or updating the final branch.
  • Worktree or branch collision: preserve the existing target and choose a new unused path or regenerated branch suggestion.
  • Replay conflict: retain recoverable task changes and ask for direction when the correct resolution is ambiguous.
  • Review or verification failure: leave the detached task worktree for continuation; do not package a commit as ready.
  • Commit-hook mutation: invalidate affected evidence, fix the missing hook-parity preflight or generated-file normalization, and repeat the required gates.
  • Non-clean or multi-commit final state: do not hand off as complete until corrected without discarding user-owned work.

© openai, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 3 other files (scripts) in .agents/skills/implementation-kickoff of openai/openai-agents-python.

  • SKILL.md
  • agents/openai.yaml
  • scripts/test_validate_handoff.py
  • scripts/validate_handoff.py

Open the folder on GitHubat commit 71c2da4

Used in 1 other repository

We found 1 copy of this SKILL.md (exact, near-identical or edited) in other folders. This page covers the copy in openai/openai-agents-python, which our catalogue first saw on October 7, 2026.

Compare with similar skills

Implementation Kickoff next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Implementation Kickoff compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Implementation Kickoff this skillopenai/openai-agents-python30k—~2.9kAutomated safety check: PassMIT
DevSpace Manual QA SetupWaishnav/devspace5.2k—~440Automated safety check: PassMIT
Nagentdavidondrej/skills4.1k—~1.7kAutomated safety check: NotesMIT
Live Extension UI Automationqixing-jk/all-api-hub4.9k—~2.6kAutomated safety check: PassAGPL-3.0
Worktree PortsFranciscoMoretti/chat-js1.2k—~231Automated safety check: NotesApache-2.0
Audit Depsgarfiec/Librechat-Mobile110—~2.4kAutomated safety check: NotesMIT

Similar skills

  • DevSpace Manual QA Setup

    Waishnav/devspace

    Prepares the current DevSpace checkout or worktree for isolated local manual QA, covering QA state seeding, UI asset builds and snapshot resets.

    5.2k GitHub stars~440 tokensUpdated 2 days ago
    DevelopmentAuto-check passed
  • Nagent

    davidondrej/skills

    Launch a new bb worker thread with the right project, model, worktree, and task brief.

    4.1k GitHub stars~1.7k tokensUpdated today
    DevelopmentAuto-check: notes
  • Live Extension UI Automation

    qixing-jk/all-api-hub

    Control, debug, and test the live dev browser extension UI (Options, Popup, Sidepanel) via CDP with persistent login states and accounts.

    4.9k GitHub stars~2.6k tokensUpdated today
    DevelopmentAuto-check passed
  • Worktree Ports

    FranciscoMoretti/chat-js

    Discover and use stable per-app ports assigned to a local monorepo worktree.

    1.2k GitHub stars~231 tokensUpdated today
    DevelopmentAuto-check: notes
  • Audit Deps

    garfiec/Librechat-Mobile

    Audit open dependabot PRs in this repo. An agent skill from garfiec/Librechat-Mobile.

    110 GitHub stars~2.4k tokensUpdated today
    DevelopmentAuto-check: notes
  • Codex CLI

    kortix-ai/suna

    Drive OpenAI's Codex CLI (codex exec) as a non-interactive coding sub-agent from inside Claude Code.

    20k GitHub stars~1.6k tokensUpdated today
    Agent WorkflowsAuto-check passed

More from openai/openai-agents-python

All 14 skills in this repo
  • Implementation Final Review

    openai/openai-agents-python

    Official

    Review completed implementation changes before final verification.

    30k GitHub stars~2k tokensUpdated today
    Auto-check passed
  • Sensitive Logging Audit

    openai/openai-agents-python

    Official

    Audit or fix sensitive-data exposure in Python SDK diagnostics, exceptions, logging, and telemetry.

    30k GitHub stars~1k tokensUpdated today
    Auto-check passed
  • Final Release Review

    openai/openai-agents-python

    Official

    Assess a Python SDK release candidate or release plan against the previous release and recommend ship or block.

    30k GitHub stars~5.4k tokensUpdated today
    Auto-check passed
  • Release Candidate Prep

    openai/openai-agents-python

    Official

    Prepare a local Python SDK release candidate in a dedicated worktree.

    30k GitHub stars~4.9k tokensUpdated today
    Auto-check passed
  • Code Change Verification

    openai/openai-agents-python

    Official

    Run the required final formatting, lint, type, and test checks after eligible SDK changes pass review.

    30k GitHub stars~1.4k tokensUpdated today
    Auto-check passed
  • Examples Run Analysis

    openai/openai-agents-python

    Official

    Analyze logs and source from a completed manual examples run.

    30k GitHub stars~1.1k tokensUpdated today
    Auto-check passed

Works with

Categories

Questions about Implementation Kickoff

What does Implementation Kickoff do?

Carry implementation through an isolated worktree and local handoff. Implementation Kickoff is an agent skill from openai/openai-agents-python, published by the product's own GitHub organization. Carry implementation through an isolated worktree and local handoff.

When should I use Implementation Kickoff?

Implementation Kickoff fits situations like: tasks that involve Git worktrees.

How do I install Implementation Kickoff in Claude Code?

Run `npx skills add openai/openai-agents-python --skill implementation-kickoff -a claude-code`. Or copy the skill folder (.agents/skills/implementation-kickoff in openai/openai-agents-python) into .claude/skills/implementation-kickoff in your project. Claude Code loads it when a task matches its description.

How do I install Implementation Kickoff in Codex?

Run `npx skills add openai/openai-agents-python --skill implementation-kickoff -a codex`. Or copy the skill folder (.agents/skills/implementation-kickoff in openai/openai-agents-python) into .agents/skills/implementation-kickoff in your project. Codex loads it when a task matches its description.

Can I use Implementation Kickoff in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add openai/openai-agents-python --skill implementation-kickoff -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/implementation-kickoff, .gemini/skills/implementation-kickoff, .github/skills/implementation-kickoff and .opencode/skills/implementation-kickoff in your project.

What does Implementation Kickoff need to run?

Going by SKILL.md and its folder, Implementation Kickoff needs Python for the scripts in its folder and the command-line tools its instructions call (git and python). Our summary lists: Python 3.

Does Implementation Kickoff access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Implementation Kickoff safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Implementation Kickoff use?

Implementation Kickoff is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Implementation Kickoff use?

About 2.9k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Implementation Kickoff?

Skills that share tags, products or a category with Implementation Kickoff: DevSpace Manual QA Setup (Waishnav/devspace, 5.2k stars), Nagent (davidondrej/skills, 4.1k stars), Live Extension UI Automation (qixing-jk/all-api-hub, 4.9k stars) and Worktree Ports (FranciscoMoretti/chat-js, 1.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Implementation Kickoff?

openai (a GitHub organization, an official publisher) maintains it in openai/openai-agents-python, which has 29,873 GitHub stars. The repository holds 14 skills in this directory. The repository was last updated on October 7, 2026.

Source: openai/openai-agents-python on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.