Agent skill

Shipping Methodology

by povvo in povvo/claudikins-kernel

A skill your agent uses when running claudikins-kernel:ship, preparing PRs, writing changelogs, deciding merge strategy, or handling CI failures — enforces GRFP-style iterative approval, code…

MITAuto-check: notesDevelopment

Install Shipping Methodology

skills CLI
$ npx skills add povvo/claudikins-kernel --skill shipping-methodology -a claude-code

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

GitHub CLI
$ gh skill install povvo/claudikins-kernel shipping-methodology --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/povvo/claudikins-kernel.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/shipping-methodology .claude/skills/shipping-methodology && 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
shipping-methodology
GitHub stars
128
Token cost
~3.2k tokens
SKILL.md length
773 words
Files
11 (incl. references)
Skills in repo
4
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when running claudikins-kernel:ship, preparing PRs, writing changelogs, deciding merge strategy, or handling CI failures — enforces GRFP-style iterative approval, code…

  • Works in 5 steps: Pre-Ship Review → Commit Strategy → Documentation (git-perfectionist) → …
  • Running claudikins-kernel:ship
  • SKILL.md covers When to use this skill, Core Philosophy, The Five Stages and Rationalizations to Resist, plus 10 more sections
  • Calls jq and git

What it does

Shipping Methodology is an agent skill from povvo/claudikins-kernel. Use when running claudikins-kernel:ship, preparing PRs, writing changelogs, deciding merge strategy, or handling CI failures — enforces GRFP-style iterative approval, code integrity validation, and human-gated merges

Its SKILL.md is about 3.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 11 other files, including reference files (for example `references/breaking-change-detection.md`, `references/changelog-merge-strategy.md` and `references/ci-failure-handling.md`).

It sits in Development, covering Changelog and release notes and Failing and flaky tests. The repository describes itself as: SRE thinking applied to Claude Code, based on Boris Cherny's Q&A. It enforces a strict 4-stage pipeline with gates between each step. You literally cannot skip verification. You… The licence is MIT.

When your agent uses it

  • Running claudikins-kernel:ship
  • Writing changelogs
  • Deciding merge strategy
  • Handling CI failures — enforces GRFP-style iterative approval

Example prompts

  • “/shipping-methodology”

Requirements

  • Pre-approved tools (allowed-tools): Read, Grep, Glob, Bash, Edit, AskUserQuestion, Skill, mcp__plugin_claudikins-tool-executor_tool-executor__search_tools, mcp__plugin_claudikins-tool-executor_tool-executor__get_tool_schema, mcp__plugin_claudikins-tool-executor_tool-executor__execute_code

Workflow steps

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

  1. Pre-Ship Review
  2. Commit Strategy
  3. Documentation (git-perfectionist)
  4. PR Creation
  5. Final Merge

What it can do on your machine

Read from SKILL.md and the folder at commit 8b626a4. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • Read
    • Grep
    • Glob
    • Bash
    • Edit
    • AskUserQuestion
    • Skill
    • mcp__plugin_claudikins-tool-executor_tool-executor__search_tools
    • mcp__plugin_claudikins-tool-executor_tool-executor__get_tool_schema
    • mcp__plugin_claudikins-tool-executor_tool-executor__execute_code

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • jq
    • git

    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

Shipping Methodology loads about 3.2k tokens when it runs, and up to ~17k if it reads all its reference files. Until then it costs about 59 tokens; SKILL.md has 773 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~59
When it runs · the whole SKILL.md, loaded when a task matches
~3.2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~17k

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: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NotePre-approves every shell command (allowed-tools: Bash)SKILL.md
    allowed-tools: Read, Grep, Glob, Bash, Edit, AskUserQuestion, Skill, mcp__plugin_claudikins-tool-executor_tool-exec

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 povvo/claudikins-kernel at commit 8b626a4, republished under its MIT licence (© povvo). 773 words, ~3,215 tokens.

Download SKILL.mdSave it as .claude/skills/shipping-methodology/SKILL.md (or your agent's skills folder). This skill also uses 10 other files; get the full folder from GitHub.
name
shipping-methodology
description
Use when running claudikins-kernel:ship, preparing PRs, writing changelogs, deciding merge strategy, or handling CI failures — enforces GRFP-style iterative approval, code integrity validation, and human-gated merges
allowed-tools
Read, Grep, Glob, Bash, Edit, AskUserQuestion, Skill, mcp__plugin_claudikins-tool-executor_tool-executor__search_tools, mcp__plugin_claudikins-tool-executor_tool-executor__get_tool_schema, mcp__plugin_claudikins-tool-executor_tool-executor__execute_code

Shipping Methodology

When to use this skill

Use this skill when you need to:

  • Run the claudikins-kernel:ship command
  • Prepare commit messages and PR descriptions
  • Update changelogs and documentation
  • Decide merge strategy (squash vs preserve)
  • Handle CI failures or merge conflicts
  • Validate code integrity before shipping

Core Philosophy

"Ship with confidence, not hope." - Shipping philosophy

Shipping is the final gate. Apply GRFP-style iterative workflow to every stage.

The Five Principles
  1. Gate check first - claudikins-kernel:verify must have passed. No exceptions.
  2. Code integrity - Ship exactly what was verified. No sneaky changes.
  3. GRFP everywhere - Section-by-section approval at every stage.
  4. Human decides - No auto-merging. Human approves final merge.
  5. Clean up after - Delete branches, update docs, celebrate.

The Five Stages

Stage 1: Pre-Ship Review

Show what's being shipped. Human confirms ready.

CheckWhat to Show
Verification statusAll phases PASS from claudikins-kernel:verify
Branches to mergeList all execute/task-* branches
Evidence summaryScreenshots, curl responses from catastrophiser
Code deltaLines added/removed, files changed

Checkpoint:

Ready to ship?

Verified: ✓ All checks passed
Branches: 3 branches to merge
Changes: +450 / -120 lines

[Continue] [Back to Verify] [Abort]
Stage 2: Commit Strategy

Draft commit message(s). Human approves.

Decision:

How should we commit?

[Squash into single commit] [Preserve commit history]

Squash (recommended for features):

  • Single clean commit on main
  • Clear feature boundary
  • Easier to revert if needed

Preserve (for large multi-part work):

  • Keeps granular history
  • Better for debugging
  • Use for multi-feature batches

Checkpoint:

Commit message:

feat(auth): Add authentication middleware

- JWT token validation
- Role-based access control
- Session management

Closes #42

[Accept] [Revise] [Back]
Stage 3: Documentation (git-perfectionist)

Update README, CHANGELOG, version. GRFP-style.

git-perfectionist uses github-readme plugin:

  1. Deep-dive on current docs
  2. Identify gaps from changes
  3. Pen-wielding for updates
  4. Section-by-section approval

Files to update:

FileWhat to Update
README.mdFeatures, usage, installation if changed
CHANGELOG.mdAdd entry in Keep a Changelog format
package.json / Cargo.toml / pyproject.tomlVersion bump if needed

Checkpoint:

Documentation updates:

README.md:
- Added: Authentication section
- Updated: Installation (new deps)

CHANGELOG.md:
- Added: v1.2.0 entry

Version: 1.1.0 → 1.2.0 (minor)

[Accept] [Revise] [Skip]
Stage 4: PR Creation

Draft PR title and body. Section-by-section approval.

PR Title Pattern:

feat(scope): Short description

PR Body Structure:

markdown
## Summary

[2-3 bullet points of what changed]

## Changes

[Detailed breakdown]

## Testing

[How it was verified]

## Screenshots

[If applicable]

Checkpoint:

PR ready to create:

Title: feat(auth): Add authentication middleware

Body:
## Summary
- Added JWT token validation
- Implemented role-based access control
...

[Create PR] [Revise] [Back]
Stage 5: Final Merge

CI passes. Human approves. Merge and cleanup.

CI Status Check:

CI Status: ⏳ Running...

[Wait for CI] [View logs] [Merge anyway]

On CI pass:

CI Status: ✓ All checks passed

Ready to merge PR #42 to main?

[Merge] [Request review first] [Cancel]

After merge:

  • Delete feature branches (unless --no-delete-branch)
  • Update ship-state.json
  • Celebrate

Final output:

Done! Shipped to main.

PR #42 merged ✓
Branches cleaned up ✓
Version: 1.1.0 → 1.2.0

Nice work!

Rationalizations to Resist

Agents under pressure find excuses. These are all violations:

ExcuseReality
"Verify passed yesterday, close enough"Stale verification = no verification. Re-run claudikins-kernel:verify.
"Just a tiny fix after verify, no big deal"Any change after verify invalidates it. Re-run claudikins-kernel:verify.
"CI is flaky, I'll merge anyway"Flaky CI hides real failures. Fix or explicitly skip with caveat.
"It's just a typo, skip the PR"All changes go through PR. No exceptions.
"Both reviewers passed, auto-merge is fine"Human approves final merge. Always.
"I'll update the changelog later"Changelog is part of shipping. Do it now.
"Force push is fine, it's my branch"Never force push to protected branches. Ever.
"Skip docs, nobody reads them"Docs are part of shipping. GRFP them.
"The gate check is too strict"The gate exists for a reason. Pass it properly.

All of these mean: Follow the methodology. Shortcuts create incidents.

Show full SKILL.md (314 more words)Show less

Red Flags — STOP and Reassess

If you're thinking any of these, you're about to violate the methodology:

  • "Let me just push this one change..."
  • "Verify is probably still valid"
  • "CI will pass next time"
  • "I'll do the PR properly next time"
  • "Auto-merge saves time"
  • "Docs can wait"
  • "Force push will fix it"
  • "The human already approved something similar"

All of these mean: STOP. Gate check exists. Human decides. Follow the stages.

Gate Check Pattern (CRITICAL)

The ship-init.sh hook enforces the claudikins-kernel:verify gate with code integrity validation.

Required Checks
bash
# 1. Verify state exists
if [ ! -f "$VERIFY_STATE" ]; then
  echo "ERROR: claudikins-kernel:verify has not been run"
  exit 2
fi

# 2. Unlock flag is set
UNLOCK=$(jq -r '.unlock_ship // false' "$VERIFY_STATE")
if [ "$UNLOCK" != "true" ]; then
  echo "ERROR: claudikins-kernel:verify did not pass or was not approved"
  exit 2
fi
Code Integrity (C-5, C-7)

Ensure we ship exactly what was verified:

bash
# C-5: Commit hash validation
VERIFY_COMMIT=$(jq -r '.verified_commit_sha' "$VERIFY_STATE")
CURRENT_COMMIT=$(git rev-parse HEAD)
if [ "$VERIFY_COMMIT" != "$CURRENT_COMMIT" ]; then
  echo "ERROR: Code changed since verification"
  exit 2
fi

# C-7: File manifest validation
VERIFIED_MANIFEST=$(jq -r '.verified_manifest' "$VERIFY_STATE")
CURRENT_MANIFEST=$(sha256sum "$MANIFEST_FILE" | cut -d' ' -f1)
if [ "$VERIFIED_MANIFEST" != "$CURRENT_MANIFEST" ]; then
  echo "ERROR: Source files changed after verification"
  exit 2
fi

If integrity check fails:

Code has changed since verification.

Verified commit: abc123
Current commit:  def456

[Re-run claudikins-kernel:verify] [Abort]

Commit Message Patterns

Feature
feat(scope): Short description

- Bullet point of what changed
- Another change

Closes #123
Bug Fix
fix(scope): Short description

Root cause: What was wrong
Fix: What we did

Fixes #456
Breaking Change
feat(scope)!: Short description

BREAKING CHANGE: What breaks and migration path

- Change 1
- Change 2
Chore (no user-facing change)
chore(scope): Short description

- Internal refactor
- Dependency update

See commit-message-patterns.md for full guide.

CHANGELOG Format

Follow Keep a Changelog format:

markdown
## [Unreleased]

## [1.2.0] - 2026-01-17

### Added

- Authentication middleware with JWT support (#42)

### Changed

- Updated error messages for clarity

### Fixed

- Token refresh race condition (#38)

## [1.1.0] - 2026-01-10

...

See changelog-merge-strategy.md for merging entries.

External Service Failures

GitHub API Failure (E-6, E-7, E-8)
PR creation failed.

Error: GitHub API returned 503

Retry 1/3...

Pattern: Max 3 retries with exponential backoff.

If persistent:

GitHub API unavailable after 3 retries.

[Try again] [Save as draft PR] [Manual merge]

Save state with pending_merge: true for recovery.

CI Failure
CI failed.

Failed checks:
- lint: 2 errors
- test: 1 failure

[View logs] [Fix and retry] [Skip CI] [Abort]

See ci-failure-handling.md for recovery patterns.

Force Push Protection (S-23)

Never force push to protected branches.

bash
# Check if target is protected
PROTECTED_BRANCHES="main master release"
if echo "$PROTECTED_BRANCHES" | grep -q "$TARGET"; then
  if [ "$FORCE" = "true" ]; then
    echo "ERROR: Cannot force push to $TARGET"
    exit 2
  fi
fi

See force-push-protection.md for safeguards.

Breaking Change Detection (S-24)

Detect breaking changes before shipping:

SignalIndicates
Removed public functionBreaking
Changed function signatureBreaking
Removed config optionBreaking
Changed default behaviourPotentially breaking

If breaking change detected:

Breaking change detected!

- Removed: authenticate() function
- Changed: login() now requires email

This requires a MAJOR version bump.

[Acknowledge and continue] [Abort]

See breaking-change-detection.md for detection patterns.

Anti-Patterns

Don't do these:

  • Shipping without claudikins-kernel:verify passing
  • Modifying code after verification
  • Force pushing to main/master
  • Auto-merging without human approval
  • Skipping documentation updates
  • Ignoring CI failures
  • Using vague commit messages ("fix stuff")
  • Forgetting to delete feature branches

State Tracking

ship-state.json
json
{
  "session_id": "ship-2026-01-17-1130",
  "verify_session_id": "verify-2026-01-17-1100",
  "target": "main",
  "phases": {
    "pre_ship_review": { "status": "APPROVED" },
    "commit_strategy": { "status": "APPROVED", "strategy": "squash" },
    "documentation": {
      "status": "APPROVED",
      "files_updated": ["README.md", "CHANGELOG.md"]
    },
    "pr_creation": { "status": "CREATED", "pr_number": 42 },
    "merge": { "status": "MERGED", "sha": "abc123" }
  },
  "cleanup": {
    "branches_deleted": ["execute/task-1-auth", "execute/task-2-routes"]
  },
  "shipped_at": "2026-01-17T11:46:00Z"
}

References

Full documentation in this skill's references/ folder:

© povvo, 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 10 other files (references) in skills/shipping-methodology of povvo/claudikins-kernel.

  • SKILL.md
  • references/breaking-change-detection.md
  • references/changelog-merge-strategy.md
  • references/ci-failure-handling.md
  • references/commit-message-patterns.md
  • references/deployment-checklist.md
  • references/force-push-protection.md
  • references/gate-failure-recovery.md
  • references/grfp-checklist.md
  • references/message-generation-fallback.md
  • references/pr-creation-strategy.md

Open the folder on GitHubat commit 8b626a4

Compare with similar skills

Shipping Methodology 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.

Shipping Methodology compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Shipping Methodology this skillpovvo/claudikins-kernel128—~3.2kAutomated safety check: NotesMIT
PR Babysitteropeninterpreter/openinterpreter69k3 repos~4.2kAutomated safety check: PassApache-2.0
Simple Englishmoeru-ai/airi50k2 repos~4.6kAutomated safety check: PassMIT
StarRocks Release NotesStarRocks/starrocks12k—~1.9kAutomated safety check: NotesApache-2.0
Cutting A ReleaseTriliumNext/Trilium38k—~3.2kAutomated safety check: PassAGPL-3.0
Systematic Debuggingultralisp/ultralisp25851 repos~2.4kAutomated safety check: PassNone

Similar skills

  • 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.

    69k GitHub starsUsed in 3 repos~4.2k tokens
    DevelopmentAuto-check passed
  • Simple English

    moeru-ai/airi

    Write or rewrite technical text with the rules of ASD-STE100 Simplified Technical English so it is clear, unambiguous, and free of AI slop.

    50k GitHub starsUsed in 2 repos~4.6k tokens
    DevelopmentAuto-check passed
  • StarRocks Release Notes

    StarRocks/starrocks

    Drafts English release notes for a StarRocks patch release from the PRs merged into its release branch, then opens a documentation PR and hands translation to /translate.

    12k GitHub stars~1.9k tokensUpdated today
    DevelopmentAuto-check: notes
  • Cutting A Release

    TriliumNext/Trilium

    A skill your agent uses when cutting, preparing, or debugging a Trilium release — bumping the monorepo version, tagging, or diagnosing a failed "Release" workflow run.

    38k GitHub stars~3.2k tokensUpdated today
    DevelopmentAuto-check passed
  • Systematic Debugging

    ultralisp/ultralisp

    A skill your agent uses when encountering any bug, test failure, or unexpected behavior, before proposing fixes

    258 GitHub starsUsed in 51 repos~2.4k tokens
    DevelopmentAuto-check passed
  • Runbook for assessing and executing a Mole CLI release: distribution channels, pre-flight checks, capital-V tags, build artifacts and the handoff to curated release notes.

    70k GitHub stars~2.6k tokensUpdated today
    DevelopmentAuto-check passed

More from povvo/claudikins-kernel

  • Brain Jam Plan

    povvo/claudikins-kernel

    A skill your agent uses when running claudikins-kernel:outline, brainstorming implementation approaches, gathering requirements iteratively, structuring complex technical plans, or facing analysis…

    128 GitHub stars~2k tokensUpdated 5 mo ago
    Auto-check passed
  • Git Workflow

    povvo/claudikins-kernel

    A skill your agent uses when running claudikins-kernel:execute, decomposing plans into tasks, setting up two-stage review, deciding batch sizes, or handling stuck agents — enforces isolation…

    128 GitHub stars~3.4k tokensUpdated 5 mo ago
    Auto-check: notes
  • Strict Enforcement

    povvo/claudikins-kernel

    A skill your agent uses when running claudikins-kernel:verify, checking implementation quality, deciding pass/fail verdicts, or enforcing cross-command gates — requires actual evidence of code…

    128 GitHub stars~2.7k tokensUpdated 5 mo ago
    Auto-check: notes

Categories

Questions about Shipping Methodology

What does Shipping Methodology do?

A skill your agent uses when running claudikins-kernel:ship, preparing PRs, writing changelogs, deciding merge strategy, or handling CI failures — enforces GRFP-style iterative approval, code…. Shipping Methodology is an agent skill from povvo/claudikins-kernel.

When should I use Shipping Methodology?

Shipping Methodology fits situations like: running claudikins-kernel:ship; writing changelogs; deciding merge strategy; handling CI failures — enforces GRFP-style iterative approval.

How do I install Shipping Methodology in Claude Code?

Run `npx skills add povvo/claudikins-kernel --skill shipping-methodology -a claude-code`. Or copy the skill folder (skills/shipping-methodology in povvo/claudikins-kernel) into .claude/skills/shipping-methodology in your project. Claude Code loads it when a task matches its description.

How do I install Shipping Methodology in Codex?

Run `npx skills add povvo/claudikins-kernel --skill shipping-methodology -a codex`. Or copy the skill folder (skills/shipping-methodology in povvo/claudikins-kernel) into .agents/skills/shipping-methodology in your project. Codex loads it when a task matches its description.

Can I use Shipping Methodology 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 povvo/claudikins-kernel --skill shipping-methodology -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/shipping-methodology, .gemini/skills/shipping-methodology, .github/skills/shipping-methodology and .opencode/skills/shipping-methodology in your project.

What does Shipping Methodology need to run?

Going by SKILL.md and its folder, Shipping Methodology needs the command-line tools its instructions call (jq and git). Its frontmatter pre-approves these tools: Read, Grep, Glob, Bash, Edit, AskUserQuestion, Skill, mcp__plugin_claudikins-tool-executor_tool-executor__search_tools, mcp__plugin_claudikins-tool-executor_tool-executor__get_tool_schema, mcp__plugin_claudikins-tool-executor_tool-executor__execute_code.

Does Shipping Methodology 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 Shipping Methodology safe to install?

Our automated static check of SKILL.md found notes only (pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Shipping Methodology use?

Shipping Methodology 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 Shipping Methodology use?

About 3.2k tokens (SKILL.md is roughly 13k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 14k tokens, read only when the agent opens those files.

What are the alternatives to Shipping Methodology?

Skills that share tags, products or a category with Shipping Methodology: PR Babysitter (openinterpreter/openinterpreter, 69k stars), Simple English (moeru-ai/airi, 50k stars), StarRocks Release Notes (StarRocks/starrocks, 12k stars) and Cutting A Release (TriliumNext/Trilium, 38k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Shipping Methodology?

povvo (a GitHub user) maintains it in povvo/claudikins-kernel, which has 128 GitHub stars. The repository holds 4 skills in this directory. The repository was last updated on April 22, 2026.

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