From the-space-memory
Interactive setup wizard to create or update tsm.toml configuration.
How this skill is triggered — by the user, by Claude, or both
Slash command
/the-space-memory:setupThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Create or update `tsm.toml` in the user's project root via step-by-step questions.
Create or update tsm.toml in the user's project root via step-by-step questions.
Check if tsm.toml already exists in $CLAUDE_PROJECT_DIR
Ask each question one at a time using AskUserQuestion. Do NOT dump all options at once.
First, scan the workspace to suggest directories:
find "$CLAUDE_PROJECT_DIR"/.. -maxdepth 2 -name "*.md" -type f 2>/dev/null | sed 's|/[^/]*$||' | sort -u | head -20
Then ask:
Which directories should be indexed? I found these candidates with .md files: (list directories found)
You can pick from these or specify your own paths. Separate multiple paths with commas.
Ask:
When the embedder (vector search) is down, how should search behave?
- error — Refuse to search (default, recommended)
- fts_only — Fall back to text-only search
Ask:
Should Claude Code session history be indexed for search? (This lets you search past conversations)
- yes (default) — Index sessions with weight 0.3
- no — Skip session indexing
Generate tsm.toml from the answers, write it, and suggest:
tsm rebuild --force # Initial index build
tsm doctor # Verify installation
# Root directory containing workspaces
index_root = "/workspaces"
# Behavior when embedder is down: "error" or "fts_only"
# search_fallback = "error"
[index]
content_dirs = [
{ path = "/workspaces/notes", weight = 1.0, half_life_days = 90.0 },
]
[index.claude_session]
weight = 0.3
half_life_days = 30.0
weight — Scoring weight, higher = more important (default: 1.0)half_life_days — Time decay half-life in days (default: 90.0)Guides collaborative design exploration before implementation: explores context, asks clarifying questions, proposes approaches, and writes a design doc for user approval.
Creates structured, bite-sized implementation plans from specs or requirements before writing code. Useful for breaking down multi-step tasks into testable steps with file structure and task boundaries.
Synthesizes the current conversation into a structured spec (PRD) and publishes it to the project issue tracker with a ready-for-agent label, without interviewing the user.
npx claudepluginhub yutaro1985/the-space-memory --plugin the-space-memory