en un clic
decision-capture
// Capture patine (decision wisdom) at Gates when KO or challenge occurs
// Capture patine (decision wisdom) at Gates when KO or challenge occurs
Iterate until success or limit, composing existing skills with Jidoka integration
Generate and maintain BPMN 2.0 diagrams linked to Gherkin scenarios
Review code changes against hard rules and conventions
Research best-in-class products using Browser MCP and WebSearch
Audit Value Delivery compliance after PR push
Systematic debugging with MCP integration, auto-invoke from qa-commit, Phase 7 Harden
| name | decision-capture |
| description | Capture patine (decision wisdom) at Gates when KO or challenge occurs |
Lightweight skill for capturing the "patine" - the accumulated wisdom of why decisions were made and alternatives rejected. Triggered at Gates when human provides KO or challenges a proposal.
| Trigger | Context | Type |
|---|---|---|
| Gate 1 KO | Wireframe rejected | Decision |
| Gate 2 KO | Scope option rejected | Decision |
| Gate 3 KO/BLOCK | Phasing challenged | Decision |
| Gate 4 KO | Technical approach rejected | Decision |
| Gate 5 Changes | PR review feedback | Decision |
| Debug Pattern | Recurring issue found | Kaizen |
| DIG 3+ times | Same wireframe refined repeatedly | Kaizen |
| Same error fixed 2+ times | Fix pattern emerged | Kaizen |
| User corrects assumption | AI was wrong about something | Hansei |
| Takt warning exceeded | Phase took longer than target | Hansei |
| Jidoka Tier 2/3 | Escalation to human required | Hansei |
Categorize the decision:
| Type | Signal | Example |
|---|---|---|
| Technical | Code, architecture, library | "Don't use GraphQL subscriptions" |
| UX | Interaction, visual, flow | "Sidebar navigation, not top nav" |
| Process | Workflow, phasing, priority | "Ship auth before tables" |
Ask for brief rationale (keep it light):
I'll note this decision for future reference.
**In one sentence, why this decision?**
Examples:
- "We tried X in 2024, broke production"
- "Users missed this in testing"
- "Conflicts with our caching strategy"
(Press Enter to skip if you prefer not to explain)
If human declines: Record decision without rationale (still valuable).
Capture metadata:
| Field | Source |
|---|---|
| Who | Current user (from context) |
| When | Current date |
| Domain | From branch name or changed files |
| Gate | Which Gate triggered capture |
| Related Task | Notion task ID if available |
Use Notion MCP to create record:
API-create-page:
parent: { database_id: "[DECISION_PATINE_DB_ID]" }
properties:
Title: { title: [{ text: { content: "[Decision summary]" }}]}
Domain: { select: { name: "[domain]" }}
Type: { select: { name: "[Technical/UX/Process]" }}
Decision: { rich_text: [{ text: { content: "[What we decided]" }}]}
Rationale: { rich_text: [{ text: { content: "[Why]" }}]}
Rejected: { rich_text: [{ text: { content: "[What we didn't do and why]" }}]}
Impact: { select: { name: "[Low/Medium/High]" }}
Gate: { select: { name: "[Gate 1/2/3/4/5/Debug]" }}
If Impact = High or Type = Technical with cross-domain effect:
ls docs/decisions/ | wc -l/docs/decisions/NNN-[slug].mdADR Template:
# ADR-[NNN]: [Title]
**Date**: [YYYY-MM-DD]
**Status**: Accepted
**Domain**: [domain]
**Captured at**: Gate [N]
## Context
[1-2 sentences: What problem were we solving?]
## Decision
[What we chose to do]
## Rationale
[Why this approach - the positive case]
## Rejected Alternatives
### [Alternative Name]
**Why not**: [Reason]
## Consequences
- [Trade-off 1]
- [Trade-off 2]
## References
- Notion: [link to Decision Patine record]
- Task: [link to related task if applicable]
For small code-level decisions during implementation:
// ADR: [Brief decision]. [Why not alternative]. —@[initials] [YYYY-MM]
Example:
// ADR: No useMemo here - profiling showed <1ms gain, adds complexity. —@mc 2026-01
Output confirmation:
**Noted:** [Decision summary]
Stored in Decision Patine database.
[If ADR created: Created ADR-[NNN] in /docs/decisions/]
Continuing with workflow...
Automatic learning capture without user prompts. These triggers capture patterns and reflections silently.
These captures happen automatically without prompting user:
Kaizen (pattern emerged):
Hansei (reflection):
Type: KAIZEN
Source: [Phase] [Loop/Commit]
Learning: "[What pattern emerged]"
Category: [UX_PATTERN | TECHNICAL | PROCESS]
Impact: LOW | MEDIUM | HIGH
Example:
Type: KAIZEN
Source: DIVERGE Loop 3
Learning: "Invite modals benefit from email preview side panel"
Category: UX_PATTERN
Impact: MEDIUM
Type: HANSEI
Source: [Phase] [Loop/Commit]
Learning: "[What we learned from the mistake/delay]"
Category: [ASSUMPTION | COMPLEXITY | PROCESS]
Impact: LOW | MEDIUM | HIGH
Example:
Type: HANSEI
Source: CONVERGE
Learning: "Original scope too ambitious - exceeded 40min takt warning"
Category: COMPLEXITY
Impact: LOW
Instead of immediately creating Notion entries for each:
Before proposing new patterns, query existing decisions:
API-query-database:
database_id: "[DECISION_PATINE_DB_ID]"
filter:
property: "Domain"
select:
equals: "[current domain]"
Use results to:
DON'T:
DO:
This skill is invoked by:
ask.mdc - Gates 1, 2, 3 on KOplan.mdc - Gate 4 on option rejectionpush-pr.mdc - Gate 5 on changes requesteddebug - When pattern worth remembering is foundDatabase: Decision Patine
| Property | Type | Required |
|---|---|---|
| Title | Title | Yes |
| Domain | Select | Yes |
| Type | Select | Yes |
| Decision | Rich Text | Yes |
| Rationale | Rich Text | No |
| Rejected | Rich Text | No |
| Challenged By | Person | No |
| Date | Date | Yes |
| Impact | Select | Yes |
| Gate | Select | No |
| Related Task | Relation | No |
| ADR File | URL | No |
Invoked automatically at Gates on KO, or manually with "use decision-capture skill".