Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/jsagir/mindrian-os-plugin/llms.txt

Use this file to discover all available pages before exploring further.

When you need to see or set the active job-to-be-done for this room, /mos:jtbd shows the per-room signal that tells Larry what kind of work you are doing right now. The JTBD state is not just a label — it is the routing signal every subsequent /mos: command reads to orient itself. A validate-idea job sharpens Bono’s grid toward the risky assumption; a find-problem job opens up the Beautiful Question walk. The heuristic classifier updates the JTBD signal silently per turn, but /mos:jtbd lets you inspect what Larry believes the job is, override it when the classifier mis-routed, and audit the full transition history. Most users never need this command directly. When you do, it is usually because Larry is reasoning from the wrong job.

​
Usage

/mos:jtbd [set <jtbd> | clear | list | history] [--json]

/mos:jtbd

​
What happens

1

Intent parsing

Larry inspects the invocation and branches: no args shows current state, list shows all 13 taxonomy entries, history shows the last 20 transitions, set <jtbd> performs a validated manual override, set with no arg renders a Shape F.1 picker, and clear returns the state to null.
2

Script execution

Larry executes the command via Bash:
node "${CLAUDE_PLUGIN_ROOT}/scripts/jtbd-command.cjs" $ARGUMENTS
The script resolves the active room from ~/MindrianRooms/.rooms/registry.json, loads <roomDir>/.mindrian/jtbd-state.json, and loads lib/hmi/jtbd-taxonomy.json for taxonomy lookups.
3

Output rendering

The script outputs a Shape E (Action Report). Larry displays the script’s stdout directly — no re-formatting, no stripped ANSI codes. Zone 4 of the report surfaces three grounded follow-up commands tuned to the active job.

​
The 13 JTBD taxonomy entries

The full taxonomy lives in lib/hmi/jtbd-taxonomy.json. All 13 entries:
IDDescription
decide-pursueDecide if a venture / opportunity / approach is worth pursuing
find-problemFind the problem worth solving — not the symptom, the root
understand-marketUnderstand the market for this idea — size, structure, dynamics
find-bottleneckFind the lagging component (Hughes 1983 reverse salient) blocking progress
prepare-pitchPrepare for an investor / partner / customer meeting
validate-ideaTest if an idea / hypothesis / claim holds up under scrutiny
compare-optionsCompare alternatives across criteria — choose between A/B/C
connect-domainsFind cross-domain analogues — what does this look like in another field?
surface-contradictionNotice and resolve internal contradictions in the room
plan-executionPlan the next 90 days — milestones, owners, dates
file-meetingCapture and route meeting intelligence into the room
audit-roomAudit Data Room health — gaps, drift, evidence tier, decision quality
exploreNo specific job — general thinking, talking, ranging (fallback)

​
Subcommands

​
Default — show current state

/mos:jtbd
Renders Shape E with the current JTBD, confidence score, evidence chain, and last 5 history entries. Example output: 
-- mindrianos -- jtbd -- find-bottleneck --

  ■ Current               find-bottleneck
     entered  2026-05-01T10:42:00Z (3m ago)
     confidence  0.82
     evidence  tokens:rs-fetch, operator:METHODOLOGY, recency:rs-thesis
     expires  manual override active until 2026-05-02T10:42:00Z

  ■ Last 5 history
     ├─ explore           2026-05-01T10:30:00Z  trigger=session_start
     ├─ explore           2026-05-01T10:35:00Z  trigger=user_message
     └─ find-bottleneck   2026-05-01T10:42:00Z  trigger=manual_set

  Summary: total transitions=3, history_used=3/50

  ▶ /mos:act                  # run methodology for active job
  ▷ /mos:jtbd set             # manual override (F.1 picker)
  ▷ /mos:jtbd history         # full transition history

​
list — browse all 13 jobs

/mos:jtbd list
Renders Shape E with all 13 taxonomy entries — each with its one-line description. Use this to find the right ID before running set.

​
set <jtbd> — manual override

/mos:jtbd set find-problem
Validates <jtbd> against the taxonomy. On success, writes the state with trigger=manual_set and a 24-hour expires_at window. On failure (unknown JTBD), prints a 3-line error to stderr and exits 1.

​
set — open Shape F.1 picker

/mos:jtbd set
Renders a Shape F.1 Next Move selector listing all 12 first-class jobs plus Free-Text. When you type a natural language selection (“set to find-bottleneck”), Larry interprets and re-invokes with the explicit ID.

​
clear — return to null

/mos:jtbd clear
Clears the current JTBD to null. Writes a history row with trigger=manual_clear. The heuristic classifier may re-populate the state on the next user turn.

​
history — full transition log

/mos:jtbd history
Renders Shape E with the last 20 transitions in chronological-recent-first order. Each row shows from → to, trigger type, and timestamp.

​
How the JTBD signal is used

The active JTBD is the first thing Larry reads before rendering any methodology surface. Every /mos: command that has a serves_jtbd list checks the active signal at startup and adjusts its behaviour accordingly: 
/mos:jtbd set validate-idea → /mos:bono (grid sharpens toward risky assumptions)
/mos:jtbd set find-problem  → /mos:beautiful-question (Why walk opens wider)
/mos:jtbd set audit-room    → /mos:status (health check surfaces gap analysis)
The heuristic classifier updates the signal silently on each turn. /mos:jtbd is the manual override when the classifier mis-routes.
The JTBD state file is per-room, never global: <roomDir>/.mindrian/jtbd-state.json. Manual overrides are sticky for 24 hours. The classifier runs silently on every turn and may update the state between your sessions.

​
Example

/mos:jtbd list
You scan the taxonomy and pick validate-idea. Then: 
/mos:jtbd set validate-idea
Larry confirms: You’re on validate-idea. Manual override active for 24 hours. Subsequent commands — including /mos:bono, which reads the active JTBD to sharpen its research grid toward risky assumptions — now orient to the validation job. 
/mos:jtbd
Shows the current state, confidence score, and the evidence chain the classifier used — useful for debugging when a command behaved unexpectedly.

/mos:analyze-needs

Score customer jobs by importance versus satisfaction — the needs analysis counterpart to the JTBD signal.

/mos:bono

The active JTBD orients Bono’s research grid — run after setting the right job.

/mos:beautiful-question

Reframe the problem before the JTBD signal routes subsequent commands.

/mos:map-unknowns

Map confident-but-thin assumptions — pairs with validate-idea JTBD to scope the debate grid.

Build docs developers (and LLMs) love

Get started for freeTalk to us