agentleFS
laaielichterlaaielichter/Reverseport/skills/port-verify/SKILL.md

port-verify

Skill18 starsChanged 7 days agoMITView on GitHub

The finish line for a ReversePort port — run port_verify, fix leftovers, missing functions and signals, PORT markers and partial units, then hand over with clear run instructions and an honest list of what still needs a human. Use before calling any port done or when the user asks whether the port is complete.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

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

    Run port_verify again until it passes (or the remaining items are explained to the user).

    line 34
  2. Distinctive:

    Never claim it compiles or plays the same unless you ran it in the target engine.

    line 46

≈ “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

Verify a port — nothing is "done" before it passes

port_verify {planId}

passed:true requires: every unit done or deliberately skipped, no missing output files, no source-engine API left in ported code (leftovers), every source function and signal present in the port (missingSymbols; lifecycle renames like _ready → Awake are understood), no units marked partial, and no dormant logic (dormant): a name that exists but has an empty or stub body, a function the source called that nothing in the port calls, a signal the port only declares, or a suppression (as any, @warning_ignore, #pragma warning disable) the port added.

Lost your context? port_verify and port_status return resume: the plan, where it stands, the next unit with its ported dependencies, and the open findings. Read it before continuing.

Fix loop

FindingFix
not ported yetport_unit {planId, unit} and port it, or port_write {status:"skipped", notes} with a reason the user accepts.
source-engine API left in the portReplace each listed pattern with the use it names, port_write again.
functions/signals not foundPort them, or rename to the matching idiom. If the function truly became unnecessary, say why in notes.
lifecycle callbacks without a counterpartFine when the work moved (e.g. connections replaced by OnTriggerEnter) — check it did.
marked partialFinish it or explain what is missing and who needs to do it.
PORT: markersEach is an open item: resolve it or list it in the hand-over.
behaviour paritytick-moved: put motion back in the physics callback. delta-dropped/added: match the source's delta use. seed-missing: seed the same way. order-dropped: map execution order/deferral. timescale-dropped: set the equivalent setting. Prove it with port_trace.
dormant logicempty-body/thin-body: port the statements. unwired: call or connect it where the source did. suppression: remove it and port the real fix.

Run port_verify again until it passes (or the remaining items are explained to the user).

Godot output: also re_apicheck {path: <output folder>, engine: "godot@4.x"}. Every renamed-* is a Godot 3 name (replace with the given Godot 4 name), every unknown-* an invented or misspelled API (use the suggestion or the real API), too-new a class the target version does not have yet.

Hand-over (short and factual)

  • Where the output is and how to open/run it in the target engine.
  • What converted automatically, what you ported by hand, what was skipped and why.
  • Open items: PORT markers, partial units, assets to re-import (models, materials, animations), settings to recreate (input actions, tags, layers), and tuning to check in play (physics rates, speeds after unit conversion).
  • Never claim it compiles or plays the same unless you ran it in the target engine.

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.