agentleFS
rodrigo-guerraarodrigo-guerraa/roblox-hub/skills/vendored/roblox-release-operations/SKILL.md

roblox-release-operations

Skill2 starsChanged 31 days agoMITView on GitHub

Run Roblox release operations for games kept in git and synced to Studio with Rojo - pre-publish gates (specs, lint, fresh-Play smoke test through the Roblox Studio MCP, saves, purchases, Studio-only guards), release notes, rollback plans for code and save-schema changes, live issue response and post-release monitoring. Use before publishing an update, when judging version readiness, or when a live release goes wrong.

1 other project uses a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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:

    Verify, and report each item:

    line 16
  2. Distinctive:

    Keep manual recovery steps for grant/data issues (reading a live save needs an Open Cloud key with datastore read scope, held by the human).

    line 43

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Roblox Release Operations

Where this conflicts with the roblox-lessons skill, roblox-lessons wins.

Goal

Ship updates with confidence and a clear recovery path. The repo is the source of truth for code; the place file holds Studio-made art; publishing from Studio is what ships, and the human owner decides when to publish and pushes commits.

Pre-release gate

Verify, and report each item:

  • Specs pass and lint is clean (selene, StyLua, type checks); no new warnings.
  • The work is committed in git; Studio is Rojo-synced to that commit (spot-check a changed script's Source), and the place is saved so place-file art is current.
  • Fresh Play through the MCP (start_stop_play): get_console_output free of errors and missing-object warnings; the core loop smoke test passes, observed with execute_luau and screen_capture, not assumed.
  • Save/load checked: an old save migrates, a new player starts clean, nothing renames the DataStore.
  • Purchases are server-authoritative: one receipt handler, PurchaseId dedupe in the save, fake-receipt tests pass.
  • Studio-only conveniences (join grants, force-owned passes, test runner) are guarded by RunService:IsStudio().
  • Product and pass ids are real and display prices match the dashboard; paid-random-item odds and policy gating still hold.
  • UI checked at phone size; performance risks known.
  • Release notes match the build.
  • Anything untestable here (real purchases, phones, multi-client) is handed to the human as a short, specific test.

Release notes

  • New content.
  • Balance changes.
  • Bug fixes.
  • Known issues when necessary.
  • Return hook or event deadline.

Rollback thinking

  • Know which changes touch the save schema. Migrations run step by step on load and must never be irreversible without a fallback.
  • During a rolling update an old server can load a save written by a newer build: it must leave that save untouched and ask the player to rejoin, never overwrite it.
  • Code rollback = revert the commit, let Rojo sync, verify, and have the human republish. A previous place version can also be restored from the place's version history, but saves already migrated forward stay migrated, so the old code must cope (see above).
  • Stage risky economy or persistence changes behind server flags where possible.
  • Keep manual recovery steps for grant/data issues (reading a live save needs an Open Cloud key with datastore read scope, held by the human).

Post-release watch

  • Error spikes.
  • Purchase/grant failures.
  • Save/load failures and kicked-for-newer-save reports.
  • Funnel drop-offs.
  • Economy inflation.
  • Community reports.

Live issue response

Reproduce in Studio first, fix in the repo, verify with a fresh Play, then the human republishes. Never hot-edit the Studio copy as the fix.

Output

  • Release readiness status.
  • Blocking issues.
  • Smoke test plan (the MCP steps) and the human-only tests.
  • Rollback or mitigation plan.
  • Post-release watchlist.

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.