From agent-smith
Refresh the latest schema for the active agent, compare it against the installed config, and produce an implementation plan for new features to adopt or deprecated settings to remove. Use when asked to review schema updates, upgrade settings, leverage new schema features, deprecate old settings, or turn schema drift into an actionable implementation plan.
How this skill is triggered — by the user, by Claude, or both
Slash command
/agent-smith:upgrade-settingsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Refresh the active agent schema, inspect the installed config against that schema, and produce an implementation plan for adopting new features and removing deprecated settings.
Refresh the active agent schema, inspect the installed config against that schema, and produce an implementation plan for adopting new features and removing deprecated settings.
Before running any scripts, resolve AGENT_SMITH_ROOT:
scripts/refresh-schemas.sh, scripts/validate-agent-config.sh, and one of .claude-plugin/plugin.json, .codex-plugin/plugin.json, opencode-plugin/package.json, or .pi/extensions/agent-smith/index.ts, use the current repo root.claude inside Claude Code, codex inside Codex, opencode inside OpenCode, and pi inside Pi. Unless the user explicitly asks for cross-agent planning, only inspect that agent family.AGENT_SMITH_TOOL=<initiating-agent> before invoking the helper scripts so they do not guess when multiple agents are installed.bash "${AGENT_SMITH_ROOT}/scripts/refresh-schemas.sh" --tool <initiating-agent>.bash "${AGENT_SMITH_ROOT}/scripts/validate-agent-config.sh" --tool <initiating-agent> --refresh and capture the output.~/.config/agent-smith/schemas/ for the initiating agent. Use the metadata file beside it to mention when the schema was fetched.~/.claude/settings.json, ~/.claude/settings.local.json, and repo-local .claude/settings.json when present~/.codex/config.toml~/.config/opencode/opencode.json~/.pi/agent/settings.json and repo-local .pi/settings.json when presentProduce a report in this format:
# Settings Upgrade Plan -- <date>
## Summary
2-3 sentences on the most important schema-driven updates.
## New Features Worth Adopting
- **Key**: schema path
- **Config file**: exact file to change
- **Why now**: why this is useful given the existing config
- **Implementation step**: smallest safe change
## Deprecations and Removals
- **Key**: schema path
- **Config file**: exact file to change
- **Current state**: how it is configured today
- **Migration**: remove, rename, or replace
## Investigate Later
- items that are newly available but not clearly justified yet
## Implementation Plan
1. Ordered steps with exact files to edit
2. Validation command(s) to rerun
3. Risks or follow-up checks
Investigate Later instead of forcing an adoption recommendation.Searches MemPalace before answering questions about past work, people, projects, or prior decisions. Returns verbatim stored content instead of guessing from model memory.
Guides Payload CMS config (payload.config.ts), collections, fields, hooks, access control, APIs. Debugs validation errors, security, relationships, queries, transactions, hook behavior.
Implements vector databases with Pinecone, Weaviate, Qdrant, Milvus, pgvector for semantic search, RAG, recommendations, and similarity systems. Optimizes embeddings, indexing, and hybrid search.
npx claudepluginhub chetmancini/agent-smith --plugin agent-smith