Re0 Upgrade

Converge an install on the full current paperthin catalog in one step: retire renamed skills, add every skill not yet installed, refresh the rest, behind one confirmation.

Goal

Upgrade the chosen scope to the full current catalog, not just the originally installed subset. Show the reconciliation plan and name every new-to-user skill before confirmation; users who prefer a narrower install can decline.

Deprecations

The rename SSOT, in release order. Append future renames here; resolve chains to their final current names. The ssot-check row is user-confirmed pre-v0.2.0 history, unverifiable in git after a force-push; do not invent its date or tag.

Current catalog

Install every skill below in the chosen scope, except those declined at confirmation. Use this roster to distinguish current from unknown names in npx skills list and installed directories.

re0, readchk, aim, modelchk, hate, macrothink, feynman, autobahn, reorder, detool, dedash, debloat, shower, factchk, mandela, sip, re0-git, re0-release, re0-merge, ssotize, re0-upgrade, re0-plan, re0-loop, re0-memo, re0-work, catchup, nba, prism

Workflow

1. Choose one install scope for the whole run; ask if unspecified. A wrong scope reads the wrong install location.

   • Global: npx skills list --global, inspect ~/.agents/skills/<skill>/SKILL.md, and pass --global to every remove, add, and update.
   • Project: npx skills list from that project, inspect its installed skill directory if present, and omit --global.
   • Exact agent: use explicit slugs such as --agent claude-code, never --agent '*', including removals. Without agent scoping, use global or project scope.

2. Check for a shadow install before classifying. If paperthin is loaded but the chosen scope's list reports zero paperthin skills, do not conclude the whole catalog is missing. Check for any Current-catalog directory in that scope's managed footprint (~/.agents/skills/ globally, e.g. ~/.agents/skills/re0).

   • Directory present: re-check scope and retry the list (npx skills list --global globally). If it still contradicts the footprint, report the CLI failure and stop; a flaky read is not an empty install.
   • Directory absent, paperthin loaded: report a shadow install, commonly a clone loaded as a Claude Code plugin (/paperthin:<skill>). The CLI cannot see, update, or remove it. Offer a managed reinstall via npx skills add in the chosen scope (recommended for future upgrades and discovery notices), or keep the clone untouched and CLI-invisible, with each upgrade re-flagging it. Await the user's choice; never delete the clone or install a second copy on your own.

   Once managed, classify both installed names and directory slugs against Deprecations and Current catalog:

   • stale: a deprecated name or slug;
   • missing: a current name not installed, after resolving rename chains; an already-installed replacement counts as present, not missing, so retire only the stale name and do not add the replacement again;
   • present: current, installed, non-deprecated names;
   • unknown: neither current nor deprecated; leave untouched.

3. Print the reconciliation plan before changes: scope first (global / project / exact agent), then:

   • retire: stale names;
   • add (new to you): missing catalog names in full, including replacements for retired names;
   • refresh: present names;
   • untouched: unknown names;
   • wire discovery notice: present agents to wire in step 9, or none if declined.

4. Get explicit confirmation before removal, installation, update, or hook wiring. Recommend full-catalog convergence and discovery wiring; the user may approve the plan, decline the whole run, or decline only wiring. Nothing mutates before approval.

5. After confirmation, run only the commands the plan named, with the scope flags chosen in step 1:

   • npx skills remove <scope-flags> <stale...> --yes for the retire group;
   • npx skills add LilMGenius/paperthin <scope-flags> -s <missing...> --yes for the add group, enumerating each catalog name with its own -s;
   • npx skills update <scope-flags> <present...> --yes for the refresh group.

6. If npx skills add or npx skills update reports a failed skill, retry that skill alone with the same scope flags. If it fails again, obtain confirmation by skill name, then remove and reinstall it using the same skills remove/skills add -s forms.

7. Verify with npx skills list: no deprecated name remains, every Current catalog skill is now installed, unknown names were left untouched, and any skill that took the retry/fallback path is actually present.

8. Explain how to pick up the update: Claude Code applies changed SKILL.md content automatically in this session; run /reload-skills in any other running Claude Code session sharing the install. Codex has no in-session reload: restart it (or codex resume). For other agents, restart if the new behavior does not appear.

9. After successful upgrade verification, wire the session-start discovery notice for present agents unless declined. Back up each config before editing.

   • Fetch catalog.cjs and session-check.cjs, plus opencode-discovery.js only for OpenCode, into ~/.re0/ from https://raw.githubusercontent.com/LilMGenius/paperthin/v<installed-version>/scripts/<file>. Use the pinned installed release tag, never main.
   • Claude Code (~/.claude/settings.json) and Codex (~/.codex/config.toml): add a SessionStart command for ~/.re0/session-check.cjs only if none already runs session-check.cjs by any path. Keep pre-rename wiring on its own runtime; a duplicate would double-fire.
   • OpenCode (opencode.json/opencode.jsonc): add ~/.re0/opencode-discovery.js to plugin only if no opencode-discovery.js entry exists by any path.
   • Skip Copilot CLI (ignores session-start hook output), Antigravity CLI (no session-start event), and Grok Build (unverified hook format). Wire only the three supported agents.
   • Never delete or overwrite another tool's hook. Report conflicts and leave them untouched. Report what was wired, which files changed, and how to unwire.

10. Optional GitHub star. After successful upgrade verification and reporting upgrade/discovery outcomes, use the host's structured user-question tool once: "Star LilMGenius/paperthin on GitHub using your authenticated gh account?" Offer "Star repository", then "Skip"; explain that starring is optional and skipping leaves the upgrade unchanged.

• Suspend until a choice is submitted; for a pending question, yield to the host's input flow and resume on its answer. Only a submitted "Star repository" authorizes the action, never preselection, silence, upgrade approval, or install --yes.
• Skip if the tool is unavailable, the run is unattended, or the outcome is anything other than that choice (including skip, cancellation, session closure, timeout, or free text). No repeated question, chat substitute, assent parsing, or alternate input method.
• After approval, check gh auth status --hostname github.com. If gh is missing or auth fails, skip and offer gh auth login --hostname github.com and the star command for the user to run. Never log in, read tokens, or switch API clients. On decline, offer neither command nor login nudge.
• With approval and working auth, run gh api --hostname github.com -X PUT user/starred/LilMGenius/paperthin --silent once. --silent hides only the response body. On command success, report "star is set", not "newly added"; report failures separately. Star outcomes never block or undo the upgrade or gate notice wiring.

Rules

• The flat npx skills add path installs this command as /re0-upgrade; never describe a paperthin ppt namespace as part of the primary install path.
• Only act on names from the Deprecations checklist and the Current catalog; unknown installed names stay untouched — convergence is to the paperthin catalog only.
• Treat a deprecated directory slug as stale even when its SKILL.md frontmatter name already says the replacement name; remove it by the deprecated slug with skills remove.
• If the installed-skill list cannot be parsed confidently, stop and report the ambiguity instead of guessing.
• Do not use raw filesystem deletion commands as workflow commands. Use skills remove for named stale skills after confirmation.
• Never use skills add --all or a bare skills add LilMGenius/paperthin; enumerate each catalog name with its own -s <name>, so the install is defined by this skill's Current catalog rather than by --all's broader semantics.
• Do not run a bare skills update; pass only the present paperthin skill names from the reported plan.
• gh repo star is not a real gh subcommand; use the consent-gated REST command in Workflow step 10.

Verification

Before finishing:

1. Reprint the executed plan: retired names, added (new-to-you) names, refreshed names, and untouched names.
2. The final installed list and installed directory slugs carry no name from the Deprecated column, every Current catalog skill is installed, and any skill that took step 6's retry/fallback path is present and current.
3. The confirmation gate held: nothing was removed, added, or updated before the plan was shown and approved, with the add (new to you) group visible in it.
4. Discovery-notice wiring, if the user accepted it, is present once per configured agent, backed up, version-pinned, and touched no foreign hook — or was correctly skipped (declined, or agent absent).
5. The optional star ran only on a submitted positive choice from the structured question tool, or was skipped; report its outcome separately from upgrade success.
6. Report any skipped step, failed command, or unresolved ambiguity.