Finishing a Development Branch
obra/superpowers
Walks the last step of a branch: confirm tests pass, detect the git environment, ask how to integrate, carry out your choice and clean up the worktree.
Behavior-first planning, verification, debugging, review, and checkpointing for non-trivial DodoStream changes.
Install Slice Development
$ npx skills add DodoraApp/DodoStream --skill slice-development -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install DodoraApp/DodoStream slice-development --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/DodoraApp/DodoStream.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/slice-development .claude/skills/slice-development && rm -rf skills-srcUse ~/.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/
Install the "slice-development" agent skill from https://github.com/DodoraApp/DodoStream/tree/main/.agents/skills/slice-development into .claude/skills/slice-development/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "slice-development", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/DodoraApp/DodoStream/tree/main/.agents/skills/slice-developmentType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add DodoraApp/DodoStream --skill slice-development -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install DodoraApp/DodoStream slice-development --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/DodoraApp/DodoStream.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/slice-development .agents/skills/slice-development && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "slice-development" agent skill from https://github.com/DodoraApp/DodoStream/tree/main/.agents/skills/slice-development into .agents/skills/slice-development/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "slice-development", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add DodoraApp/DodoStream --skill slice-development -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install DodoraApp/DodoStream slice-development --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/DodoraApp/DodoStream.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/slice-development .cursor/skills/slice-development && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "slice-development" agent skill from https://github.com/DodoraApp/DodoStream/tree/main/.agents/skills/slice-development into .cursor/skills/slice-development/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "slice-development", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/DodoraApp/DodoStream.git --path .agents/skills/slice-development--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add DodoraApp/DodoStream --skill slice-development -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install DodoraApp/DodoStream slice-development --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/DodoraApp/DodoStream.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/slice-development .gemini/skills/slice-development && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "slice-development" agent skill from https://github.com/DodoraApp/DodoStream/tree/main/.agents/skills/slice-development into .gemini/skills/slice-development/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "slice-development", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install DodoraApp/DodoStream slice-developmentInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add DodoraApp/DodoStream --skill slice-development -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/DodoraApp/DodoStream.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/slice-development .github/skills/slice-development && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "slice-development" agent skill from https://github.com/DodoraApp/DodoStream/tree/main/.agents/skills/slice-development into .github/skills/slice-development/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "slice-development", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add DodoraApp/DodoStream --skill slice-development -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install DodoraApp/DodoStream slice-development --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/DodoraApp/DodoStream.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/slice-development .opencode/skills/slice-development && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "slice-development" agent skill from https://github.com/DodoraApp/DodoStream/tree/main/.agents/skills/slice-development into .opencode/skills/slice-development/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "slice-development", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
slice-development- GitHub stars
- 243
- Token cost
- ~1.9k tokens
- SKILL.md length
- 814 words
- Files
- 1
- Skills in repo
- 7
- Repo updated
- First seen
- Licence
- GPL-3.0
At a glance
Behavior-first planning, verification, debugging, review, and checkpointing for non-trivial DodoStream changes.
- Works in 10 steps: INTAKE — confirm the outcome, affected… → SLICE_PLAN — record the selected slice… → DEFINE_EXPECTATION — state the exact… → …
- Development work in your project
- SKILL.md covers Inputs, State loop, Verification ladder and Checkpoint policy, plus 1 more section
- Calls pnpm
What it does
Slice Development is an agent skill from DodoraApp/DodoStream. Behavior-first planning, verification, debugging, review, and checkpointing for non-trivial DodoStream changes.
Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Development. The licence is GPL-3.0.
When your agent uses it
- Development work in your project
Example prompts
- “/slice-development”
Workflow steps
10 steps, taken from the first numbered list in SKILL.md.
- INTAKE — confirm the outcome, affected platforms, number of behavioral slices, assigned worktree, and permission boundary.
- SLICE_PLAN — record the selected slice and its existing callers, owner, RED expectation, proof path, and checkpoint subject.
- DEFINE_EXPECTATION — state the exact observable behavior before changing code.
- ESTABLISH_RED — use a focused failing Jest test for deterministic utilities, parsers, stores, reducers, hooks, transforms, and API…
- IMPLEMENT_MINIMUM — change only the causal code and the tests/flow needed for this slice. Reuse existing repository primitives and leave…
- TARGETED_VERIFY — run the lowest ladder level that can observe the expectation. Escalate only when a changed boundary is not observable at…
- DEVICE_REQUIRED? — if the expectation involves native Android/TV behavior, use android-interactive-verification only with explicit…
- SLICE_REVIEW — inspect the slice for unrelated files, unnecessary abstractions, duplicate or stale paths, weak assertions, debug noise…
- CHECKPOINT_COMMIT — commit the reviewed slice before selecting the next one. A conversation checkpoint is not a Git checkpoint.
- INTEGRATION_VERIFY — after the last slice, run the union of affected checks and the final gate, then compare the complete change against…
What it can do on your machine
Read from SKILL.md and the folder at commit 1ef9461. 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
Shell commands in SKILL.md call:
pnpm
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use pnpm, 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
Slice Development loads about 1.9k tokens when it runs. Until then it costs about 32 tokens; SKILL.md has 814 words of instructions outside code blocks.
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); files beside SKILL.md are not scanned.
SKILL.md
The full file from DodoraApp/DodoStream at commit 1ef9461, republished under its GPL-3.0 licence (© DodoraApp). 814 words, ~1,949 tokens.
Download SKILL.mdSave it as
.claude/skills/slice-development/SKILL.md (or your agent's skills folder).- name
- slice-development
- description
- Behavior-first planning, verification, debugging, review, and checkpointing for non-trivial DodoStream changes.
Slice development
Use this skill for a non-trivial behavior change. It is mandatory when autonomous work contains two or more behavioral slices. A slice is one user-visible or system-observable outcome plus the smallest coherent state, UI, API, and test work required to deliver it. Do not split a behavior into technical layers such as types, service, component, and tests.
Inputs
Before editing, identify:
- The requested outcome and affected platforms.
- The existing tests, flows, fixtures, utilities, and components that own the behavior.
- Every known caller of an exported symbol that will change; use LSP references when available.
- Whether device verification is authorized and whether an agent-device MCP session is available.
- Already-created checkpoints and any contracts they establish.
Write no more than one concise record per slice:
Goal: <single user or system outcome>
Slice <n> — <observable behavior, not implementation layer>
Expectation: <user/system can ...>
RED: <focused failing test command, or reproducible device interaction + expected evidence>
GREEN scope: <smallest named files/symbols; existing utility/component to reuse>
Verify: <L0–L6 command(s) and expected observation>
Device: no | conditional <specific trigger> | yes <native boundary>
Checkpoint: <Conventional Commit subject, or “single-slice final commit”>State loop
- INTAKE — confirm the outcome, affected platforms, number of behavioral slices, assigned worktree, and permission boundary.
- SLICE_PLAN — record the selected slice and its existing callers, owner, RED expectation, proof path, and checkpoint subject.
- DEFINE_EXPECTATION — state the exact observable behavior before changing code.
- ESTABLISH_RED — use a focused failing Jest test for deterministic utilities, parsers, stores, reducers, hooks, transforms, and API behavior. For native focus, visual, lifecycle, player, or layout behavior, use a repeatable interaction and expected hierarchy, screenshot, or log evidence; never invent a low-value unit test.
- IMPLEMENT_MINIMUM — change only the causal code and the tests/flow needed for this slice. Reuse existing repository primitives and leave the app coherent.
- TARGETED_VERIFY — run the lowest ladder level that can observe the expectation. Escalate only when a changed boundary is not observable at that level.
- DEVICE_REQUIRED? — if the expectation involves native Android/TV behavior, use
android-interactive-verificationonly with explicit authorization and an agent-device MCP session. - SLICE_REVIEW — inspect the slice for unrelated files, unnecessary abstractions, duplicate or stale paths, weak assertions, debug noise, accidental snapshots, and behavior outside the stated expectation.
- CHECKPOINT_COMMIT — commit the reviewed slice before selecting the next one. A conversation checkpoint is not a Git checkpoint.
- INTEGRATION_VERIFY — after the last slice, run the union of affected checks and the final gate, then compare the complete change against the original request.
If a check fails, enter DEBUG: rerun the same reproduction, narrow the failure using the relevant diff/log/hierarchy, state one causal hypothesis, change one causal area, and rerun the targeted check. After three consecutive attempts without new evidence, return to SLICE_PLAN and split or replace the remaining slice, add diagnostic evidence, or escalate only if the current level cannot observe it. Completed checkpoints remain valid unless their contract is disproven. Revise only unstarted slice records when understanding changes.
Show full SKILL.md (384 more words)Show less
Verification ladder
Select the first level that observes the stated behavior; higher levels are for newly exposed boundaries, not ceremony.
| Level | Execute | Use for |
|---|---|---|
| L0 — static | pnpm typecheck and pnpm exec eslint <changed-file>... | Every TypeScript slice before checkpoint; imports, types, hooks, and token names. |
| L1 — focused unit | pnpm test -- <test-file> or pnpm test -- -t "<name>" | Pure parsers, reducers/stores, utilities, transforms, and API error mapping. |
| L2 — focused integration/component | The owning hook, RNTL component, database, or API test file | Rendered interactions and React Query/Zustand, SQLite, or mocked-network boundaries. |
| L3 — affected family/build | The narrow affected directory/package and required non-device build | Cross-component/package boundaries, fixture changes, remote UI, or Expo/native plugin generation. |
| L4 — interactive Android/TV | The authorized agent-device MCP session | Native focus, D-pad, player/lifecycle/system behavior, real recycling, Android defects, and rendered geometry. |
| L5 — targeted Maestro | The named flow with stable selectors and reviewed baselines | Important deterministic regression paths after interactive behavior is understood. |
| L6 — final affected gate | pnpm verify:ci plus affected package/build checks and selected L5 flows | The final union after all slices, not every edit. |
Skip L4/L5 for pure logic and mockable behavior. Enter L4 before claiming behavior involving Platform.isTV, TVFocusGuideView, hasTVPreferredFocus, nextFocus*, useTVEventHandler, real LegendList recycling, native players, immersive/system UI, lifecycle, or Android/TV geometry. Run L5 only when L4 establishes a stable, important, selector- and baseline-backed regression path. Never start Metro or use Expo web for the release-APK Android E2E route.
Checkpoint policy
For autonomous multi-slice work, each checkpoint contains exactly one observable behavior and everything required for it: production code, focused tests or a reviewed flow/baseline, translations, and tracked generated content when applicable. It passes L0 and its selected targeted proof and uses <type>(<scope>): <imperative summary>.
Amend the preceding checkpoint only when all of these hold: the next slice has not started, the correction is solely required by that same slice's expectation or test, and the commit has not been reported, handed off, or used as a later baseline. Otherwise create a new fix(...) checkpoint, including for integration defects found after valid checkpoints. Do not create commits for discovery, failed experiments, formatting-only churn, or ignored generated artifacts.
Final evidence
Report only observed evidence using this literal structure:
Implemented:
- <observable behavior>
Checkpoints:
- <hash> <subject>
Verified:
- <level>: <command or interaction> — <observed result>
Not run:
- <check> — <concrete reason>
Remaining concerns:
- <actual concern, or none>Do not claim a skipped device, CI, credentialed API, screenshot, or Maestro result. Do not duplicate the repository's durable styling, focus, API, or command rules; those remain in AGENTS.md.
© DodoraApp, GPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in .agents/skills/slice-development of DodoraApp/DodoStream.
Open the folder on GitHubat commit 1ef9461
Compare with similar skills
Slice Development 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.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Slice Development this skillDodoraApp/DodoStream | 243 | — | ~1.9k | Automated safety check: Pass | GPL-3.0 | |
| Finishing a Development Branchobra/superpowers | 296k | 5 repos | ~1.9k | Automated safety check: Pass | MIT | |
| Typescript Advanced Typesrolling-scopes/rsschool-app | 10k | 24 repos | ~4.2k | Automated safety check: Pass | MPL-2.0 | |
| PR Babysitteropeninterpreter/openinterpreter | 69k | 3 repos | ~4.2k | Automated safety check: Pass | Apache-2.0 | |
| Code Review ChecklistshareAI-lab/learn-claude-code | 78k | 5 repos | ~1.1k | Automated safety check: Pass | MIT | |
| Greplooponyx-dot-app/onyx | 32k | 4 repos | ~3.3k | Automated safety check: Pass | MIT |
Similar skills
Typescript Advanced Types
rolling-scopes/rsschool-app
Master TypeScript's advanced type system including generics, conditional types, mapped types, template literals, and utility types for building type-safe applications.
PR Babysitter
openinterpreter/openinterpreter
Watches an open GitHub pull request until it merges, handling review comments, diagnosing CI failures and retrying flaky checks along the way.
Code Review Checklist
shareAI-lab/learn-claude-code
Reviews code against a five-part checklist covering security, correctness, performance, maintainability and testing, and reports findings in a fixed format.
Greploop
onyx-dot-app/onyx
Iteratively improves a PR (GitHub), MR (GitLab), or shelved changelist (Perforce) until Greptile gives it a 5/5 confidence score with zero unresolved comments.
Guidelines
akash-network/node
Behavioral guidelines to reduce common LLM coding mistakes. An agent skill from akash-network/node.
More from DodoraApp/DodoStream
Agent Verification
DodoraApp/DodoStream
Select and run concise deterministic DodoStream verification checks.
Bug Reproduction
DodoraApp/DodoStream
Reproduce a reported bug on Android/Android TV and capture evidence.
Issue Triage
DodoraApp/DodoStream
Rank open GitHub issues and propose the next work. An agent skill from DodoraApp/DodoStream.
Feature Delivery
DodoraApp/DodoStream
Deliver a feature end to end from user-visible goal to PR-ready branch.
PR Handoff
DodoraApp/DodoStream
Prepare the branch, title, and draft PR body for substantive finished work.
Android Interactive Verification
DodoraApp/DodoStream
Session-scoped Android and Android TV fixture, APK, interaction, evidence, and Maestro procedure through the agent-device MCP server.
Related
Categories
Questions about Slice Development
What does Slice Development do?
Behavior-first planning, verification, debugging, review, and checkpointing for non-trivial DodoStream changes. Slice Development is an agent skill from DodoraApp/DodoStream. Behavior-first planning, verification, debugging, review, and checkpointing for non-trivial DodoStream changes.
When should I use Slice Development?
Slice Development fits situations like: development work in your project.
How do I install Slice Development in Claude Code?
Run `npx skills add DodoraApp/DodoStream --skill slice-development -a claude-code`. Or copy the skill folder (.agents/skills/slice-development in DodoraApp/DodoStream) into .claude/skills/slice-development in your project. Claude Code loads it when a task matches its description.
How do I install Slice Development in Codex?
Run `npx skills add DodoraApp/DodoStream --skill slice-development -a codex`. Or copy the skill folder (.agents/skills/slice-development in DodoraApp/DodoStream) into .agents/skills/slice-development in your project. Codex loads it when a task matches its description.
Can I use Slice Development 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 DodoraApp/DodoStream --skill slice-development -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/slice-development, .gemini/skills/slice-development, .github/skills/slice-development and .opencode/skills/slice-development in your project.
What does Slice Development need to run?
Going by SKILL.md and its folder, Slice Development needs the command-line tools its instructions call (pnpm).
Does Slice Development access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Slice Development 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. Review the folder before installing.
What licence does Slice Development use?
Slice Development is published under the GPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Slice Development use?
About 1.9k tokens (SKILL.md is roughly 7.8k 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 Slice Development?
Skills that share tags, products or a category with Slice Development: Finishing a Development Branch (obra/superpowers, 296k stars), Typescript Advanced Types (rolling-scopes/rsschool-app, 10k stars), PR Babysitter (openinterpreter/openinterpreter, 69k stars) and Code Review Checklist (shareAI-lab/learn-claude-code, 78k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Slice Development?
DodoraApp (a GitHub organization) maintains it in DodoraApp/DodoStream, which has 243 GitHub stars. The repository holds 7 skills in this directory. The repository was last updated on September 20, 2026.
Source: DodoraApp/DodoStream on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.