agentleFS
amikaiamikai/openings-mcp/.agents/skills/release/SKILL.md

release

Skill106 starsChanged 3 months agoMITView on GitHub

Use when the user asks to cut, tag, or ship an openings-mcp release (e.g. "release v0.8.0"), to run pre-release provider smoke tests, or to rewrite a fresh release's auto-generated notes.

  • Commits and pushes

193 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use create-agent-adapter, changelog-style and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • MicrosoftSkill

    create-agent-adapter

    Technical guide for creating a new Paperclip agent adapter. Use when building a new adapter package, adding support for a new AI coding tool (e.g. a new CLI agent, API-based agent, or custom process), or when modifying the adapter system. Covers the required interfaces, module structure, registration points, and conventions derived from the existing claude-local and codex-local adapters.

    Also in 5 projects that use this

  • GrinvSkill

    changelog-style

    Turn a commit range into CHANGELOG.md entries following common-changelog style. Use when finishing a PR or cutting a release, to write/fix the [Unreleased] section's entries — moving [Unreleased] under a new dated version heading is a separate, automated step (see the release skill / scripts/sync-version.ts), not this skill's job.

    Also in 4 projects that use this

  • Microsoft AzureSkill

    component-release

    Prepares a verifiable Agent Landing Zone UI release. Use for release branches, VERSION, changelog entries, tags, release notes, and compatibility evidence; never publish without explicit human approval.

    Also in 4 projects that use this

  • GrinvSkill

    docs-consistency-check

    Check README/manifest.json/server.json/CHANGELOG.md/AGENTS.md and docs/*.md for drift against the actual registered tools and source. Use after adding, renaming, or removing a tool, or as part of a live-audit pass.

    Also in 4 projects that use this

  • Microsoft AzureSkill

    documentation-consistency

    Keeps AI Landing Zone contributor, user, and operator documentation aligned with shipped Bicep behavior. Use for parameters, defaults, outputs, topology, deployment flow, operations, or releases.

    Also in 8 projects that use this

  • parcadeiSkill

    agent-context-isolation

    Agent Context Isolation

    Also in 3 projects that use this

  • +3 more from Grinv

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 183 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Test HEAD through the real MCP path, never an installed binary or the session's connected openings-mcp server — those run the previous …

    line 27
  2. Very specific:

    Only the ## sections that apply: Highlights (a minor release's lead story, opening phrase bolded), Added, Changed, Fixed.

    line 89

“Common” is measured: kept by at least 6 in 10 of the 183 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Release

Overview

Cut a release end to end: validate the version, live-smoke-test every provider at HEAD, push the tag, wait for the Release workflow (goreleaser publishes the GitHub release with a raw commit-list body), then rewrite the notes in the house style and report back.

Input

Exactly one version matching ^v\d+\.\d+\.\d+$ (all numeric, e.g. v0.8.0). Reject anything else — 0.8.0, v0.8, v0.8.0-rc1 — and ask for a corrected version instead of guessing. It must not already exist and must sort above the latest tag (git tag --sort=-v:refname | head -1 — that latest tag is also the "previous version" used in steps 2, 4, and 5). Require a clean tree on main, in sync with origin/main.

1. Smoke-test every provider (before tagging)

Test HEAD through the real MCP path, never an installed binary or the session's connected openings-mcp server — those run the previous release. Build once (go build -o <scratchpad>/openings-mcp ./cmd/openings-mcp) and drive it over stdio; the request script is in the integrate-new-provider skill's MCP-surface step.

  • ATS adapters (the unified-search set: providerOrder in cmd/verify-companies/main.go): per provider, sample 3–5 companies from internal/provider/<name>/companies.yaml; search_jobs_by_company must return live listings for each, and get_job_detail_by_company must succeed on at least one returned job_id.
  • Dedicated-tool providers (job boards and careers sites: the Register* calls in newServer, cmd/openings-mcp/main.go): per provider, 3–5 varied <name>_search_jobs queries exercising the filters it supports (keyword, location, ...) plus one <name>_get_job_detail on a returned identifier.

A failure stops the release: retry once to rule out upstream flakiness, then report to the user instead of tagging. Also run go test ./... — cheaper to catch breakage here than across six CI matrix legs after the tag is public.

2. Tag

Annotated, message in the same shape as the release title:

git tag -a vX.Y.Z -m "vX.Y.Z: <one-line summary of the release>"
git push origin vX.Y.Z

Skim git log <prev>..HEAD --oneline first to write the summary line.

3. Wait for CI

The tag push triggers .github/workflows/release.yml: unit tests on six platforms, then goreleaser (GitHub release, archives, Homebrew cask, Docker images). Watch it to completion:

gh run watch --exit-status \
  "$(gh run list --workflow=release.yml --limit 1 --json databaseId -q '.[0].databaseId')"

On failure, report the failing job; never delete the tag or release to retry without asking.

4. Rewrite the release notes

Collect what shipped since the previous tag: git log <prev>..vX.Y.Z --oneline plus the merged PRs it references — read enough of each PR to state its user-facing impact, not just its subject line. Then replace goreleaser's title and commit-list body:

gh release edit vX.Y.Z --title "..." --notes-file <scratchpad>/notes.md

House style (see v0.5.5, v0.6.0, v0.7.0 for worked examples):

  • Title vX.Y.Z: <one-line story of the release>.
  • Only the ## sections that apply: Highlights (a minor release's lead story, opening phrase bolded), Added, Changed, Fixed.
  • Bullets are full sentences about user-visible outcomes, not commit restatements; concrete numbers (roster counts) when they carry the story; each bullet ends with its PR refs (#N) — GitHub autolinks them in release notes, no full URL needed.
  • Internal-only changes (CI, refactors, docs, skills) are omitted unless users can observe the difference.
  • Last line: **Full Changelog**: https://github.com/amikai/openings-mcp/compare/<prev>...vX.Y.Z.
  • English throughout.

5. Report

End the run by showing the user the full changelog (the compare link plus the commit/PR list gathered in step 4) and the rewritten release note verbatim.

Common Mistakes

  • Smoke-testing through a released binary or the session's MCP server — that validates the previous release, not the one being cut.
  • Tagging before the smoke tests finish: the tag push immediately triggers a public release build.
  • Writing notes from commit subjects alone — the style demands user-facing impact, which usually lives in the PR diff or description.
  • Reciting provider lists from memory: derive the ATS set and the dedicated tools from the wiring named in step 1; both change from release to release.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.