" \ --topics "<a,b,c>" --goals "<g1,g2>" --source study-session \ --confidence <n> --importance <n> --body "<body>" ``` 5. If `brain add` rejects the note, fix the draft and retry — never write files into `knowledge/` directly. **Finish every run (ux.md #2/#3/#6 — one command per tool call):** 1. `.venv/bin/python -m brain ingest`, then `.venv/bin/python -m brain graph` — the user never learns regeneration exists. 2. Snapshot: `git add` the new note, then `git commit -m "snapshot: log: <note title>"`. No git? One line: snapshot skipped, notes still saved. 3. End with the receipt block (docs/ux.md #2): the note id + topics it landed with; self-confidence stated as claimed, not proven; "map data refreshed — reload the tab"; "saved a local snapshot — nothing leaves your machine"; next action (usually: quiz this topic when you're ready to prove it).

--- name: log description: Capture a study session, project work, or learning moment as a structured note in the second-brain knowledge base. Use when the user describes something he studied, built, struggled with, or learned — even one sentence. --- # /log — capture a study session Turn the user's freeform account into a validated note. The CLI is the only write path. 1. Read `config.yaml` (domains, sources) and `goals/goals.yaml` (goal ids). 2. Check for related notes: `.venv/bin/python -m brain search "" -k 3`. Sessions are events, so still create a new dated note — but reference related existing notes in the body as `[[note-id]]` so the graph gets the edge. 3. Draft the note: - **domain**: best fit from config domains. - **topics**: 2–5 lowercase tags. Reuse the existing vocabulary where possible (`grep -rh "^topics:" knowledge/ | sort -u` or check recent notes) — topic consistency is what makes the graph connect. - **goals**: only goals the content genuinely advances (empty is fine). - **confidence** (SELF-assessed, 1–5): infer from his language — "struggled", "confused", "kept getting it wrong" → 2; "got it working", "makes sense" → 3; "comfortable", "explained it to someone" → 4. Unsure → pick the lower value. - **importance** (1–5): how much this matters to the goals it touches. - **body**: his account in his voice. Preserve struggle markers verbatim (negative evidence drives gap analysis) and concrete application details (they are the evidence for ai_confidence later). Do not polish away specifics. 4. Create it (title becomes the id slug): ``` .venv/bin/python -m brain add --domain --title "" \ --topics "<a,b,c>" --goals "<g1,g2>" --source study-session \ --confidence <n> --importance <n> --body "<body>" ``` 5. If `brain add` rejects the note, fix the draft and retry — never write files into `knowledge/` directly. **Finish every run (ux.md #2/#3/#6 — one command per tool call):** 1. `.venv/bin/python -m brain ingest`, then `.venv/bin/python -m brain graph` — the user never learns regeneration exists. 2. Snapshot: `git add` the new note, then `git commit -m "snapshot: log: <note title>"`. No git? One line: snapshot skipped, notes still saved. 3. End with the receipt block (docs/ux.md #2): the note id + topics it landed with; self-confidence stated as claimed, not proven; "map data refreshed — reload the tab"; "saved a local snapshot — nothing leaves your machine"; next action (usually: quiz this topic when you're ready to prove it).