verify-laterdog
Skill2 starsChanged 29 days agoApache-2.0View on GitHub
Prove later.dog's workspace delegation, durable supervisor, remote publishing and conversation recovery on its own isolated fixtures, keeping the evidence and naming the limits.
In this repository43 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- llms.txtllms.txt
- architectSkill
- benchmark-checklistSkill
- correct-recurring-errorsSkill
- create-verification-skillSkill
- independent-reviewSkill
- maintain-verification-skillSkill
- monitor-and-shipSkill
- orchestrate-cloud-workSkill
- setup-laterdogSkill
- verify-laterdogSkill
- architectSkill
- benchmark-checklistSkill
- correct-recurring-errorsSkill
- create-verification-skillSkill
- independent-reviewSkill
- maintain-verification-skillSkill
- monitor-and-shipSkill
- orchestrate-cloud-workSkill
- setup-laterdogSkill
- verify-appSkill
- architectSkill
- benchmark-checklistSkill
- correct-recurring-errorsSkill
- create-verification-skillSkill
- independent-reviewSkill
- maintain-verification-skillSkill
- monitor-and-shipSkill
- orchestrate-cloud-workSkill
- setup-laterdogSkill
- verify-laterdogSkill
- architectSkill
- benchmark-checklistSkill
- correct-recurring-errorsSkill
- create-verification-skillSkill
- independent-reviewSkill
- maintain-verification-skillSkill
- monitor-and-shipSkill
- orchestrate-cloud-workSkill
- setup-laterdogSkill
- verify-laterdogSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
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
- Very specific: line 8
Use after changing the supervisor, its MCP tools, the Cloud jobs screen or the hand-off back to a conversation, before saying the change …
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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.
Files
Verify later.dog
Use after changing the supervisor, its MCP tools, the Cloud jobs screen or the hand-off back to a conversation, before saying the change works. Node 24 and pnpm. Other server and conversation changes follow docs/verification/README.md.
Procedure
- Check the checkout:
pnpm typecheckandpnpm lint. - Run the fixtures:
pnpm laterdog:test. It starts its own HTTP server, SQLite store, Git repositories, React render and a disposable later.dog server on the fake engine. It never submits a real cloud task or opens a real pull request. Its conversation round trip checks that server withcontrol-laterdog doctorfirst, then leavesreceipts.jsonandserver.login.laterdog-evidence/conversation/(andconversation-pull/for the hosted wake-up path). - See it in the app.
node scripts/laterdog-preview.mjsprints a JSON line withpreviewUrl. Open it with the browser automation your client has, open Cloud jobs in the sidebar (Settings → Advanced mode shows it), connect repositoryfixture/previewwith environment IDsynthetic-environment, delegate one narrow task, and follow the job's Conversation link back to the chat that asked. The job should stay queued: the preview's scheduler is off and no provider is called, so a queued synthetic job is the expected result. Save screenshots in.laterdog-evidence/preview/, showing the scope, the job's identity, its receipts and the conversation it returned to. - Stop what you started: Ctrl-C the preview process, and only that one. It copies its server log to
.laterdog-evidence/preview/server.log, closes its servers and deletes its temporary data. Then check that the evidence is still there.
What this does not prove
Fixture success does not qualify a live provider. Codex Cloud and Claude cloud compatibility, real GitHub access, the hosted supervisor, a signed macOS build and throughput are qualified only by the live checklist in docs/laterdog/verification.md. Per-feature recipes and their limits: features/README.md.
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
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.

