tim-hub/powerball-harness / .claude/skills/release-this
release-this
VerifiedOrchestrates the full powerball-harness plugin release with build, validation, version bump, CHANGELOG, tag, and GitHub Release. Use when releasing this plugin.
Release This Plugin
Project-specific release orchestrator for the claude-code-harness plugin. Runs all plugin-specific checks (build, consistency, validation, Codex templates, version sync), then handles the full release: version bump, CHANGELOG update, commit/tag/push, and GitHub Release creation.
Quick Reference
| Argument | Behavior |
|---|---|
patch | Patch version bump (bug fixes, x.y.Z+1) |
minor | Minor version bump (new features, x.Y+1.0) |
major | Major version bump (breaking changes, X+1.0.0) |
--dry-run | Run steps 1–5 only; print what would happen; skip publishing |
--complete | Mark release complete only (step 7 — empty commit + push) |
Execution Flow
Step 1: Cross-platform binary build if there are changes in go/ since the last release
Run make build-all to compile darwin-arm64, darwin-amd64, and linux-amd64 binaries unless there are no changes in go/ source files since the last release. This ensures that any build issues are caught before proceeding with release validation or publishing steps.
make build-all
Fail immediately if make build-all exits non-zero. Do not proceed to step 2.
Step 2: Plugin consistency checks
Run the 13 plugin-specific consistency checks. These verify symlink integrity, template consistency, version alignment, and structural invariants.
bash "${CLAUDE_SKILL_DIR}/scripts/check-consistency.sh"
Step 3: Full plugin validation
bash ./tests/validate-plugin.sh
This runs all CI-gated checks: marketplace.json structure, skill frontmatter, hook schemas, section 9 migration residue scan, and section 10 skill-description audit.
Step 4: Codex template verification
Verify that the Codex CLI configuration templates and the Codex-specific skill bodies are present and readable. Missing files would ship a broken Codex experience to users.
4a. Codex CLI config templates (harness/templates/codex/):
ls -la harness/templates/codex/
Confirm AGENTS.md, README.md, config.toml, .codexignore, and rules/ are all present.
4b. Codex skill bodies (harness/templates/codex-skills/):
ls -la harness/templates/codex-skills/
ls harness/templates/codex-skills/*/SKILL.md
Confirm each subdirectory (breezing, harness-schedule-run, harness-work) contains a SKILL.md. A missing SKILL.md means that Codex-variant skill is broken.
Step 5: Version sync check
Verify that harness/VERSION and harness/harness.toml agree before proceeding.
HARNESS_RELEASE_EXTRA_VERSION_FILES="harness/harness.toml" bash "${CLAUDE_SKILL_DIR}/scripts/sync-version.sh" check
If there is a mismatch, stop and ask the user to run:
HARNESS_RELEASE_EXTRA_VERSION_FILES="harness/harness.toml" bash "${CLAUDE_SKILL_DIR}/scripts/sync-version.sh" sync
or manually reconcile the files before proceeding.
Dry-run stops here. In --dry-run mode, steps 1–5 run normally but no release is published, no files are written, and no git commits or tags are created. Report what the release would contain and what the new version would be, then exit.
Step 6: Release phases (preflight → CHANGELOG → commit/tag/push → GitHub Release)
Create the wrapper lock:
mkdir -p .claude/state && touch .claude/state/harness-release-wrapper.lock
Phase 0–2 (preflight + version bump): Delegate to releaser agent via setup invocation:
Agent(
subagent_type: "releaser",
description: "run preflight and bump version",
prompt: "{\"schema_version\":\"releaser-request.v1\",\"invocation\":\"setup\",\"bump_type\":\"<patch|minor|major>\"}"
)
Parse new_version from the releaser-response.v1 response. If status is "error", surface the error field and abort.
Phase 3 (CHANGELOG): Move [Unreleased] content into ## [X.Y.Z] - YYYY-MM-DD. Keep empty ## [Unreleased] above it. Format rules: references/writing-changelog.md.
Phase 3.5 (Validate release notes): Validate CHANGELOG and release notes:
bash "${CLAUDE_SKILL_DIR}/scripts/validate-release-notes.sh"
Phase 4–5 (commit/tag/push): Delegate to releaser agent via finalize invocation:
Agent(
subagent_type: "releaser",
description: "commit, tag, push release v<new_version>",
prompt: "{\"schema_version\":\"releaser-request.v1\",\"invocation\":\"finalize\",\"version\":\"<new_version>\"}"
)
Parse git_hash from the response. If status is "error", surface the error and changes fields.
Phase 6 (GitHub Release): Create the GitHub Release:
gh release create "v${NEW_VERSION}" \
--title "v${NEW_VERSION} - <one-line summary>" \
--notes "$(cat <<'EOF'
## What's Changed
...
EOF
)"
Wait for GitHub Release creation to succeed before proceeding to step 7.
Step 7: Completion marking commit
After step 6 finishes, create an empty commit to mark the release as fully complete, then push it. Remove the wrapper lock when done.
NEW_VERSION=$(cat harness/VERSION)
git commit --allow-empty -m "chore: mark v${NEW_VERSION} release complete"
git push origin "$(git rev-parse --abbrev-ref HEAD)"
rm -f .claude/state/harness-release-wrapper.lock
This empty commit is the explicit record that "all release work is done." It is separate from the chore: release vX.Y.Z commit created by the releaser agent.
--complete Mode
When called with --complete, execute only step 7 (the completion marking commit). Use this if step 6 already finished but the completion commit was not created.
NEW_VERSION=$(cat harness/VERSION)
# Verify a GitHub Release exists for this version before marking complete
gh release view "v${NEW_VERSION}" || { echo "ERROR: GitHub Release v${NEW_VERSION} not found. Create it first."; exit 1; }
git commit --allow-empty -m "chore: mark v${NEW_VERSION} release complete"
git push origin "$(git rev-parse --abbrev-ref HEAD)"
Related Skills
harness-review— Run before release to catch issues earlyharness-plan— Plan the next release tasksharness-work— Work on tasks
Related Rules and References
harness/rules/versioning.md— SemVer classification criteria and batch release policyharness/rules/github-release.md— GitHub Release Notes formatreferences/versioning-rules.md— Detailed versioning rulesreferences/writing-changelog.md— CHANGELOG format guide
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 470k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 197k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 197k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 197k | Updated | Claude CodeCodex | Verified | Install |