Prompt file imported from PauloHYBEX/Nexus-TCMS (
.windsurf/workflows/list-phase-assumptions.md). Copyright stays with the author.
List Phase Assumptions
Surface what Cascade intends to do for a phase before committing to a plan. This is analysis of what Cascade thinks, not intake of what you want — use discuss-phase for that.
Use when: You want to validate direction before planning, or you're unsure what approach will be taken.
Usage: list-phase-assumptions [N]
Step 1: Validate Phase
Phase number is required:
Usage: list-phase-assumptions [N]
Example: list-phase-assumptions 3
Find phase [N] in ROADMAP.md:
grep -E "Phase [N]:" .planning/ROADMAP.md
If not found: list available phases and stop.
Step 2: Load Context
cat .planning/ROADMAP.md
cat .planning/PROJECT.md
cat .planning/STATE.md
Check for CONTEXT.md (user decisions):
ls ".planning/phases/" 2>/dev/null | grep "^0*[N]-"
If CONTEXT.md exists, read it — it affects assumptions.
Step 3: Analyze Phase
Based on the phase goal, project context, and codebase, surface assumptions across five dimensions. Be honest about confidence levels.
## My Assumptions for Phase [N]: [Phase Name]
### Technical Approach
What libraries, frameworks, or patterns I'd use:
- [Fairly confident:] I'd use [X] because [reason from phase context]
- [Assuming:] I'd structure this as [Y] because it fits the existing pattern
- [Unclear:] [Z] could go multiple ways depending on [ambiguity]
### Implementation Order
What I'd build first, second, third:
1. [First thing] — because it's foundational
2. [Second thing] — because it depends on #1
3. [Third thing] — because it integrates the pieces
### Scope Boundaries
**In scope:** [What I think this phase covers]
**Out of scope:** [What I'd defer to later phases]
**Ambiguous:** [Things that could go either way]
### Risk Areas
Where I anticipate complexity or challenges:
- [Tricky thing]: [Why it's hard and what I'd watch out for]
- [Potential issue]: [What could go wrong]
### Dependencies
**From prior phases:** [What must exist before this runs]
**External:** [Third-party dependencies needed]
**Feeds into:** [What later phases will need from this phase's output]
---
**What do you think?**
Are these assumptions accurate? Tell me:
- What I got right
- What I got wrong
- What I'm missing
Wait for user response.
Step 4: Acknowledge Feedback
If corrections provided:
Key corrections noted:
- [correction 1]
- [correction 2]
This changes my approach significantly: [summarize updated understanding]
If assumptions confirmed:
Assumptions validated. Ready to plan.
Step 5: Offer Next Steps
ask_user_question([
{
header: "Next Step",
question: "What would you like to do with these assumptions?",
multiSelect: false,
options: [
{ label: "Discuss Phase", description: "Run discuss-phase [N] — you answer my questions to build CONTEXT.md" },
{ label: "Plan Now", description: "Run plan-phase [N] — create plans now (assumptions noted above apply)" },
{ label: "Re-examine", description: "Analyze again with your corrections" }
]
}
])
🛑 STOP. Wait for the user's reply before continuing.
Note: Any corrections discussed here are not automatically captured. Run discuss-phase [N] to write them to a CONTEXT.md that planners will read.
Learning Checkpoint
Read learning_mode from .planning/config.json.
If auto: Offer:
💡 Learning moment: About to commit to an approach. Log the path not taken:
@agentic-learning either-or— Record the alternatives considered, the choice made, and expected consequences. When the build goes sideways, this record tells you which assumption failed.
If manual: Add quietly: "Tip: @agentic-learning either-or to log the approach decisions being made here."