← all publishers

latent-variable

@latent-variable source repo

Report this profile Claim this profile

• 5 skills
• 0 followers
• 6 hours ago last updated

1. ▌
   Cleanup · latent-variable

   Full consistency sweep of the agent knowledge base: user scope (~/.agents) plus every repo's project scope. Finds contradictions between memories, decisions that landed in one layer and not the others, stale facts, canon restated inside repos, and project-scope overrides that no longer make sense. Fixes what it finds in the same turn. Load when the user asks for a cleanup, a canon review, or "check our scope", and after any session that changed a rule in several places.
   0
   installs
2. ▌
   Remember · latent-variable

   Save a durable fact about the user or their work into the shared cross-agent memory (~/.agents/memory), then commit + push so every agent (Claude, Codex, Gemini, Pi) sees it. The write-back path: use it whenever you learn something worth persisting beyond this session, or when the user says "remember this".
   0
   installs
3. ▌
   Onboarding · latent-variable

   First-run setup for userscope. Interview the user, fill the AGENTS.md template + memory with their identity and preferences, detect which agent CLIs are installed, wire them all to ~/.agents, and tune optional features (plain-language explanations, the Speak-to-me TTS block). Run this the first time someone clones userscope, or when they say "onboard me" / "set up userscope".
   0
   installs
4. ▌
   Self Correct · latent-variable

   Audit and repair the shared agent knowledge base (~/.agents) when it drifts from reality: removed/renamed dirs or repos, changed paths/flags/commands, superseded facts. Keeps instructions, memory, and skills true for every agent on the machine. Use on demand, on a schedule, or whenever you notice canon is stale mid-task.
   0
   installs
5. ▌
   Project Scope · latent-variable

   How agent context works at PROJECT scope (inside a repo) vs USER scope (~/.agents). Where project memory + shared skills live, how the symlink bridge works, and the scope-confusion trap to avoid. Load when working in any repo that has a `.agents/` dir, or when bootstrapping a new repo.
   0
   installs