upgrade
Agent skill fromagenticnotetaking/arscontexta
Apply plugin knowledge base updates to an existing generated system. Consults the Ars Contexta research graph for methodology improvements, proposes skill upgrades with research justification. Never auto-implements. Triggers on "/upgrade", "upgrade skills", "check for improvements", "update methodology".
How to install upgrade
install upgrade
# install the arscontexta pack with the skills CLI
npx skills add agenticnotetaking/arscontexta# or add only this skill to Claude Code: copy the folder
skills/upgrade/
# into
~/.claude/skills/upgrade/ (all projects)
.claude/skills/upgrade/ (one project)
1## Runtime Configuration (Step 0 — before any processing)23Read these files to configure domain-specific behavior:451. **`ops/derivation-manifest.md`** — vocabulary mapping, platform hints6 - Use `vocabulary.notes` for the notes folder name7 - Use `vocabulary.note` / `vocabulary.note_plural` for note type references8 - Use `vocabulary.reduce` for the extraction verb9 - Use `vocabulary.reflect` for the connection-finding verb10 - Use `vocabulary.reweave` for the backward-pass verb11 - Use `vocabulary.verify` for the verification verb12 - Use `vocabulary.rethink` for the meta-cognitive verb13 - Use `vocabulary.topic_map` for MOC references14152. **`ops/config.yaml`** — processing depth, domain context16173. **`ops/derivation.md`** — derivation state and engine version1819If these files don't exist, use universal defaults.2021## EXECUTE NOW2223**Target: $ARGUMENTS**2425Parse immediately:26- If target contains a specific skill name (e.g., "upgrade reduce"): check only that skill27- If target contains "--all": check all generated skills28- If target is empty: check all generated skills (same as --all)2930**START NOW.** Reference below defines the upgrade process.3132---3334## Why Consultation, Not Hashing3536Skills do not upgrade through hash comparison against a generation manifest. Hash comparison answers a narrow question: "Has this file changed?" Meta-skill consultation answers the right question: "Is this skill's approach still the best approach given what we know?"37…
Excerpt shown under the repository's MIT license.
