Menu
Initialize memhub for a repo in OpenCode; use when the user asks to bootstrap project memory.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Read-only health check of the memhub project in this repo — schema version, render freshness, write-log activity
Bootstrap memhub in this repo — initialize the SQLite store, seed starter state and architecture narratives, render PROJECT.md and PROJECT_LEDGER.md
Summarize this memhub session, route updates into the database, then re-render PROJECT.md and PROJECT_LEDGER.md
Rebuild + install memhub and bring every memhub instance on this machine (each known repo DB + the machine-global store) to head schema, resync installed agent skill wrappers, with a one-time fix for the ~/.local/bin PATH shadow. Run from the memhub source repo.
Wrap up an OpenCode memhub session; run the binary-rendered policy, draft updates, get per-item approval, write, and render.
Run memhub's read-only CLAUDE.md/AGENTS.md drift-and-bloat audit (`memhub audit md`) and interpret the findings — size, generated-file drift, managed-block version, keystone phrases, opt-in user-global size — recommending concrete fixes without rewriting anything itself. Trigger on: "audit CLAUDE.md", "check for CLAUDE.md drift", "is CLAUDE.md bloated", "run the md audit", "/audit-md".
| name | init-project |
| description | Initialize memhub for a repo in OpenCode; use when the user asks to bootstrap project memory. |
| compatibility | opencode |
Bootstrap memhub in the current repo, with an approval gate before writing starter narratives.
Workflow:
memhub is absent from PATH or .memhub/ already exists.git rev-parse --show-toplevel; ask before initializing anywhere else.AGENTS.md, CLAUDE.md, and other context systems.memhub init --json, then write approved bodies with memhub state set --from-file ... --actor opencode:init-project --json and memhub arch set --from-file ... --actor opencode:init-project --json.memhub render and report generated paths.mode = "hybrid" under [retrieval] in .memhub/config.toml, then memhub index rebuild --actor opencode:init-project to backfill embeddings — a rebuild is required whenever this mode changes, not just the config edit) vs. the FTS-only default (nothing to do).memhub code index to warm up memhub locate//locate now instead of on first use.memhub global enable for the shared ~/.memhub/global.sqlite store.memhub doc add "<path>" --json to make a reference doc's chunks surface in plain recall.memhub sync enable plus [sync] drive_subpath in .memhub/config.toml pointing at a folder that already syncs, then memhub sync status to confirm.AGENTS.md exists, offer a minimal one pointing at .memhub/rendered/; wait for approval before writing.