agentleFS
mewziciansmewzicians/Skills-Folder/skills/game-release/SKILL.md

game-release

Skill0 starsChanged 24 days agoApache-2.0View on GitHub

Synchronize and byte-verify maintained game release deliverables, such as a repository mirror or Pages site; run pre-publish checks; build archives or upload bundles on request. Use when the user explicitly asks to prepare, refresh, check, or hand off a public release or upload, not ship readiness.

  • Reads credentials

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run only when the user asked for this workflow.

    line 8
  2. Distinctive:

    Read repository guidance and references/project-profile.md.

    line 10
  3. Distinctive:

    Hard rule: before anything is published, check and report each item below.

    line 20
  4. Distinctive:

    Don't bend one because a request is casual: ask the user.

    line 36
  5. Distinctive:

    Don't add a rule after your own mistake or a reviewer's nitpick without that yes.

    line 53
  6. Distinctive:

    Ask about new rules once, batched at the end of the task.

    line 56
  7. Distinctive:

    When a file hits its size cap, or at a milestone, prune it: cut rules no one can tie to a reason.

    line 57

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Game Release

Run only when the user asked for this workflow; otherwise offer it.

Read repository guidance and references/project-profile.md. Identify the authoritative root, the maintained deliverables, the tested candidate version, and the public allowlist. Preserve unrelated work and stop on unexplained implementation-document drift.

Keep two kinds of output apart. Maintained deliverables, such as the public repository mirror or a store, jam, or Pages upload the project keeps current, are synchronized and verified on every release. On-demand copies, such as a ZIP archive, are built only when the user asks, and never kept, synchronized, or checked in: a kept archive drifts on every documentation edit, while every check it carries already holds against the folder it came from. A live upload, such as to Pages, a store, or a jam, is publishing.

Treat a mirror as an allowlist: synchronize only mapped root files, byte-for-byte, keep required hidden repository files, and keep launchers separate from game logic. Exclude credentials, private archives, local profiles, caches, editor and vault state, stale archives, and accidental personal paths. Run project QA checks and required change-focused tests before release.

Hard rule: never delete a stale mirror file that no mapping covers, because the mirror may hold the only copy. Report it, and act only on the user's approval naming the path and the action. In a git mirror, remove it as its own commit only when its current bytes are already committed; never commit other pending edits to get that backup. Otherwise move it to a dated quarantine folder with collision-safe names and a list of original paths. An approved removal never permits publishing the backup.

When the user asks for an archive, run python scripts/build_repository_zip.py --help, then build with it. Use --allowlist FILE for anything public: it is the only mode that refuses an unlisted file, while the default excludes are a denylist of common names and can miss a secret. Include required hidden files deliberately; the .git* exclude also drops .github/, .gitattributes, and .gitignore unless they are --included. Review the packaged and excluded lists. Report contentDigest (names and bytes) apart from the archive's own SHA-256, which is reproducible across Python builds only with --stored.

Hard rule: before anything is published, check and report each item below. Anything pushed to a public remote can be copied before you notice.

  • identity: every author, committer, tagger, and Co-authored-by trailer being published uses a noreply or project address, never a personal one, on every branch and tag being pushed. When the remote already has history, check what it lacks: git log --branches --tags --not --remotes=origin --format="%h %ae %ce %(trailers:key=Co-authored-by,valueonly,separator=%x20)". For a first publication, drop --not --remotes=origin. Check taggers with git for-each-ref refs/tags --format="%(refname:short) %(taggeremail)";
  • secrets: no tokens, private keys, credentials, or .env files in the files or the history being published;
  • paths: no absolute local paths, such as a user's home folder;
  • private folders: nothing outside the allowlist, such as notes, vaults, continuity ledgers, or local profiles;
  • licenses: every bundled asset, font, sound, and library permits redistribution, with its required notices included.

Report the checks first: the deliverables and any archive, synchronized files, each check as Pass, Fail, Not Run, or Blocked, candidate and archive digests, file count, and remaining blockers. Hard rule: never call a release ready when a required check failed or did not run.

Then stop. Hard rule: publish only on the user's explicit instruction for this candidate, given after that report, and never while a required check failed or did not run. Without that instruction the agent never pushes, tags, uploads, or publishes. Publishing, deleted untracked files, and other external side effects can't be undone with git.

Rules And Pushback

Lines marked Hard rule guard approvals, deletion, publishing, and private data, where a mistake is costly or can't be undone; each gives its reason. Don't bend one because a request is casual: ask the user. Unmarked lines still bind too. Scope limits, honest evidence, and approval steps aren't optional. Scale each step's effort to the risk, and say so when you skip one that doesn't fit. A profile or instruction file records an owner decision only where it says the owner made it; otherwise treat it as project guidance. If a rule seems wrong for this task, say which one and why before you continue.

How Rules Get Made

  1. Something the user says in passing applies to the current task only.
  2. Exception: anything about money, privacy, publishing, deleting, or permissions binds as soon as they say it. Ask once whether to make it permanent.
  3. A rule becomes standing only when the user says "make this a rule", or you ask and they say yes. Record their words, the date, and the reason with it.
  4. This covers every file that steers an agent: instruction files, style guides, skills, checklists, templates, CI checks, and agent memory.
  5. Don't add a rule after your own mistake or a reviewer's nitpick without that yes. Fix the input or the work instead.
  6. A new ruling replaces the one it contradicts. Never keep both.
  7. Ask about new rules once, batched at the end of the task.
  8. When a file hits its size cap, or at a milestone, prune it: cut rules no one can tie to a reason. The user's taste decisions are flagged for them and never cut by an agent.

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.