Files
claude-mds/TODO.md
T

87 lines
4.3 KiB
Markdown

# TODO
## Bootstrap installer (next session — 2026-08-14)
**Goal:** clone this repo on any machine, run one command, and every file lands
where the tooling expects it.
git clone <this repo> ~/projects/claude-mds && ~/projects/claude-mds/install.sh
### Shape
- **A manifest, not a script full of paths.** Declare `repo path -> target path`
in one file and let `install.sh` walk it. Adding a doc later should mean one
manifest line, not editing shell.
- **Symlinks, not copies.** Same reason as here: two copies drift, and the point
of the repo is that an edit lands in git. `ln -sfn`.
- **Idempotent.** Safe to re-run after every `git pull`. Re-linking an existing
correct link is a no-op.
- **Conditional targets.** `user/CLAUDE.md` installs everywhere. `officer.dev/`
only where that workspace exists. A manifest entry needs a "only if this path
exists" condition.
- **Never clobber silently.** If the target is a real file and not our link,
back it up (`.bak-<date>`) and say so. That file may be the only copy of
something written by hand on that machine.
- **Dry run by default, or at least `--dry-run`.** Show what it would link.
### Tools are installed, not referenced in place — DECIDED
`tools/` stays in the repo as the source, and the installer puts it under
`~/.local/claude-scripts/` (confirm the name), with `mask-secret` linked onto
`~/.local/bin` so it is a command rather than a path.
That fixes a real bug: `user/CLAUDE.md` currently hardcodes
`~/projects/claude-mds/tools/mask-secret.py`, which is only true on this
machine. Once installed, the doc says `mask-secret` and is true everywhere.
Carry `mask-hooks.d/` across with it, keeping the executable bits. The hooks are
the machine-specific half, so the installer should place only the hooks that
belong on that machine — same "only if this path exists" condition as the docs.
`officer-chat-events` should not land on a laptop with no Officer install, even
though it already exits cleanly when it finds no database.
Update `user/CLAUDE.md` to name the command in the same change, or the doc and
the layout disagree the moment this ships.
### Prior art to discuss: akitaonrails/ai-memory
<https://github.com/akitaonrails/ai-memory> — read 2026-08-13. Solves a superset
of this: cross-vendor long-term memory so you can quit Claude Code mid-task and
continue in Codex without re-explaining. A Rust binary plus a data dir, served
over MCP, with `install-mcp` and `install-hooks` subcommands.
Worth stealing even if we never run it:
- **`_global` scope.** A reserved project that is unioned into every query —
standing personal rules, tech choices, style. Exactly what `user/CLAUDE.md`
is. Confirms the split we already made.
- **Page categories**: `_rules/`, `decisions/`, `procedures/`, `gotchas/`.
Better than one growing file. Our `user/CLAUDE.md` is six sections and will
not stay readable at twenty.
- **Authority-aware retrieval.** Maintained rules outrank episodic session
evidence, but do not become absolute. A good answer to "the doc says X but
the code does Y" — which bit us today with the stale `pm2` rule.
- **`.ai-memory.toml` marker file** in a repo ancestor to pin which
workspace/project a directory belongs to. Directly relevant to our manifest's
"only if this path exists" condition — a marker is more explicit than a path
test.
- **`bootstrap`** seeds pages from `git log`, README, `docs/`, module headers.
The thing we would want when adding a new machine or repo.
The real decision: it is a **running service** (daemon, SQLite, FTS5,
embeddings, MCP, hooks). Ours is plain files and symlinks, readable with `cat`,
and works when nothing is installed. That simplicity is a feature, especially
given today's "code on disk is not what is running here" problem. Options:
adopt it, borrow the taxonomy only, or stay flat and revisit when the flat file
gets unreadable.
### Open questions
- Which files are genuinely universal vs this-machine-only? Right now
`officer.dev/CLAUDE.md` is the only machine-specific one, but the
`platform-appstore/` set is a candidate and was never folded in.
- Should `settings.json` / hooks be managed here too, or only `.md` files? The
repo is named for the docs but already carries `tools/`.
- Bootstrapping the bootstrap: the machine needs git and the repo URL before any
of this runs. Worth a one-line `curl | sh`, or deliberately not.