elicit-brain

name

elicit-brain

description

Surface and codify tacit knowledge from the user's local second-brain — knowledge implied everywhere in their transcripts and notes but written nowhere. Use this skill when the user wants to discover what they know implicitly, fill in knowledge gaps, or turn unwritten rules into authored vault notes. Backs `brain elicit` (interactive draft-then-correct loop, needs Ollama) and `brain elicit list` (Ollama-free gap queue view). For plain document lookup or Q&A use `consult-brain`; for themes and patterns across interactions use `brain-graph`. MANDATORY TRIGGERS: elicit my tacit knowledge, what do I know that I haven't written down, what haven't I written down, surface my knowledge gaps, surface my tacit knowledge, what implicit knowledge, draft the unwritten rules, what rules am I not writing down, interview me about, what do I keep referencing without writing, brain elicit, knowledge elicitation, codify what I know, fill in my knowledge gaps, what's in my head that isn't in my notes, tacit knowledge, gap in my second brain, unwritten knowledge, what am I implicitly assuming, what have I never documented.

Elicit Brain

Use brain elicit to surface tacit knowledge — things the corpus shows the user knows implicitly (referenced constantly, implied everywhere) but has never written as an authored vault note. The interaction is draft-then-correct: the system makes a confident guess at the unwritten rule and the user corrects it. The correction is the elicited knowledge. Codified notes are tagged tacit + the signal kind and include a ## Source section wikilinking the evidence documents.

For plain document lookup and Q&A, use consult-brain. For themes and patterns across interactions, use brain-graph. For action-item queries, use brain-todo.

When to use this skill

The user says…	Reach for
"What do I know that I haven't written down?"	brain elicit list then brain elicit
"Surface my knowledge gaps"	brain elicit list --json
"Interview me about [topic]"	brain elicit --target "[topic]"
"Draft the unwritten rules I keep referencing"	brain elicit --signal delta
"What implicit knowledge do I have about [person/project]?"	brain elicit --target "[name]"
"Show me my knowledge gaps without drafting anything"	brain elicit list (Ollama-free)

The gap signals

Signal	What it detects
delta	Entities heavily referenced in ingested docs (transcripts, Slack, mail) but never authored into a vault note — the cleanest proxy for tacit knowledge.
orphan	Graph entities with high mention-count but no written description. The graph knows who matters; this surfaces why it doesn't.
contradiction	Opposing positions across docs (flag-gated: BRAIN_ELICIT_CONTRADICTION_ENABLED=true; requires Ollama + non-null summaries).
user_flagged	A topic the user names with --target; always enters the queue regardless of corpus signals.

Viewing the gap queue — brain elicit list

Read-only. Runs the active detectors, upserts results, and renders the ranked open gaps. Does not need Ollama.

brain elicit list                    # refresh detectors and show the queue
brain elicit list --json             # machine-readable (for synthesis)
brain elicit list --limit 10         # show at most 10 gaps
brain elicit list --low-confidence   # include gaps below the score floor
brain elicit list --type person --type org   # only person/org gaps (repeatable)

JSON shape per gap: signal_kind, target_type, target_id, score, evidence_ids (list of doc IDs), rationale, status.

--type filters the queue to one or more entity types (person / org / project / topic / tool / doc, repeatable) — pass it when the user scopes the ask ("what haven't I written down about my projects").

Interactive loop — brain elicit

Requires Ollama. For each gap in the ranked queue the command drafts a confident rule text and opens it in $EDITOR. The user corrects/accepts; on a non-empty, changed body the result is codified as a vault note.

brain elicit                                 # run the loop over the full ranked queue
brain elicit --target "engineering culture"  # user-flagged: jump to this specific topic
brain elicit --signal delta                  # only surface delta gaps this run
brain elicit --signal orphan
brain elicit --signal contradiction          # needs BRAIN_ELICIT_CONTRADICTION_ENABLED=true
brain elicit --type project                  # only draft gaps about projects (repeatable)
brain elicit --include-low-confidence        # include below-score-floor gaps

--type (repeatable: person / org / project / topic / tool) narrows the loop to gaps about those entity types — combine with --signal to be precise ("interview me about my orgs, delta signal only").

Interactive keymap:

Key	Action
e	Open draft in $EDITOR. Save a non-empty, changed body to codify as a vault note.
s	Skip (dismiss) this gap.
n	Snooze — prompts for days, suppresses until then.
q	Quit the loop.

The editor buffer opens with the drafted rule and a comment block listing the evidence doc IDs. The body must be non-empty and changed from the draft — re-saving the draft unchanged re-prompts rather than codifying.

After codification

A codified gap produces:

• A vault note at brain note new's path, tagged tacit + signal kind.
• A ## Source footer with [[wikilinks]] to the evidence documents.
• The elicitation_gaps row updated to status='resolved' — the gap will not resurface.

Read the new note with brain show <id-prefix>. Search future sessions with brain search "<topic>" --tag tacit.

Config knobs

Env var	Default	Purpose
BRAIN_ELICIT_MIN_EVIDENCE_DOCS	3	Minimum source docs required before a gap is draftable.
BRAIN_ELICIT_MIN_GAP_SCORE	0.3	Post-normalization score floor (0–1).
BRAIN_ELICIT_QUEUE_LIMIT	20	Max gaps surfaced per list run.
BRAIN_ELICIT_CONTRADICTION_ENABLED	false	Enable the contradiction detector.
BRAIN_ELICIT_CONTRADICTION_MIN_DOCS	5	Min entity doc-count for contradiction scan.

Prerequisites check

Before running brain elicit (the loop), verify:

brain doctor        # Postgres + Ollama must be healthy
brain elicit list   # confirms the table exists and detectors run cleanly

If brain doctor reports Ollama unavailable, brain elicit list still works (Ollama-free). The interactive loop (brain elicit) will fail until Ollama is running.

When NOT to use this skill

• Plain "find docs about X", a quote, a single fact → consult-brain.
• "Themes in my conversations with X", "what connects A and B" → brain-graph.
• "What are my open action items?" → brain-todo.
• The user explicitly says "don't elicit" / "just search my brain".