How to place a bet
Commit a selected solution as the bet that anchors capability mapping.
Use this when: You have evaluated your solution options and are ready to commit the team to a chosen direction and produce the bet.md artifact that map-capabilities will reason against.
Prerequisites: product-engineering pack installed; a solutions artifact from diverge-solutions or explore-options (recommended); validation-notes.md from de-risk-intent (optional but folds kill conditions in automatically).
Result: A committed bet.md at <output_dir>/shaping/<slug>/bet.md recording the chosen option, confidence, appetite, rationale, accepted risks, and assumptions — ready to anchor step 6.
Goal: You have evaluated options and completed (or deliberately skipped)
validation. You are ready to commit the team to a chosen direction and produce
the artifact map-capabilities will reason against.
Skill: place-bet (PE pack, user scope)
De-risk this self-service export-retention approach, then place a bet if its kill condition survives.When to use place-bet vs de-risk-intent
Section titled “When to use place-bet vs de-risk-intent”de-risk-intent (step 3.5) and place-bet (step 5) are sequential, not
alternatives. Run both when the bet rests on a fragile assumption worth testing
first; skip de-risk-intent when the risk is accepted rather than worth
investigating.
This guide belongs to the robust shaping path.
| Step | Skill | Purpose |
|---|---|---|
| 3 | diverge-solutions | Generate structured comparable options |
| 3.5 | de-risk-intent | Validate the riskiest assumption before committing |
| 5 | place-bet | Commit the chosen direction; produce bet.md |
| 6 | map-capabilities | Use bet.md as the anchor for capability mapping |
Rule of thumb: if you can articulate your chosen direction and name the
risks you are accepting, place-bet is ready. If a core assumption has not
been tested and the cost of a wrong bet is high, run de-risk-intent first.
What inputs place-bet needs
Section titled “What inputs place-bet needs”The skill reads what it finds and accepts free-form otherwise:
- A solutions artifact (optional but recommended) — any
*solution*or*options*file under<output_dir>/shaping/<slug>/. This is typically the output ofdiverge-solutionsorexplore-options. If absent, the skill offers to rundiverge-solutionsand degrades to free-form if you decline. validation-notes.md(optional) — the output ofde-risk-intent, if you ran it. Read if present; never required. Kill conditions and risks found there are folded into the betting table automatically.
You do not need to have run diverge-solutions specifically. The skill accepts
options from any prior source — external research, workshop notes, informal
comparison, or an explore-options run.
How to fill each betting table field
Section titled “How to fill each betting table field”option — the chosen direction in one line. Name it specifically. “Direct download with streaming” is better than “probably the streaming one.”
option-source — where the option came from. A file path when the options
came from a structured artifact; free-form when you described them informally.
This is your traceability anchor: map-capabilities can trace back to the
options that were considered.
confidence — your honest confidence that this option will pay off: high,
medium, or low. A low-confidence bet with explicit risks-accepted is better
than a high-confidence bet that papers over uncertainty.
appetite — the time budget the team is willing to commit before reassessing. Name a number rather than “open” when possible: “3 weeks” creates a forcing function; “open” doesn’t. Appetite is not an estimate — it is the boundary you are placing on the bet.
rationale — why this option over the alternatives you considered. Reference the options by name. One short paragraph is enough.
risks-accepted — an explicit list of known risks you are taking on. Fold in
anything from validation-notes.md. The point is not to eliminate risk — it is
to name it so it can be tracked and revisited.
assumptions — what must be true for this bet to pay off. A bet with clear assumptions is easier to reassess when conditions change.
kill-condition (optional) — the signal that would reverse this decision.
Pull from de-risk-intent output when available. If not, leave it blank — the
field is optional. A kill condition is most valuable when confidence is medium
or low.
How to hand the bet to map-capabilities
Section titled “How to hand the bet to map-capabilities”When bet.md is written, the skill prints a workspace.toml transition
suggestion. Treat it as routing input: use work-intake to materialize the
canonical artifact and register schema-valid lifecycle state. Do not paste a
comment-backed or slug-only entry manually.
To hand off: run map-capabilities and point it to the committed bet at
<output_dir>/shaping/<slug>/bet.md. It will use the option, assumptions,
and next-step pointer as its anchor for the capability map.
What happens if no solutions artifact exists
Section titled “What happens if no solutions artifact exists”If no *solution* or *options* file is found at the shaping slug path, the
skill offers to run diverge-solutions first and names the impact: without
structured comparable options, the rationale and risks-accepted in the betting
table are less defensible.
You can decline and proceed with a free-form description of the options you
considered. The resulting bet.md will record option-source: free-form so
the absence of a structured artifact is visible in the audit trail.
See also
Section titled “See also”de-risk-intent— step 3.5; validates the riskiest assumption before committingdiverge-solutions— step 3; generates structured comparable optionsexplore-options— an alternative way to surface options for the betting tablemap-capabilities— step 6; usesbet.mdas the anchor for capability mappingwork-intake— records the suggested transition through canonical artifact-first intake
What you have now
Section titled “What you have now”- A committed
bet.mdat<output_dir>/shaping/<slug>/bet.mdwith the chosen option, appetite, rationale, risks, and assumptions. map-capabilitiescan now use that bet as the anchor for the next shaping step.