fpf-strategize

This is strategic betting, not variant selection. You are choosing which class of approach to bet on — scoping the search space for variant generation. "We'll explore variants within [family X] because [reason]."

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "fpf-strategize" with this command: npx skills add m0n0x41d/principled-claude-code/m0n0x41d-principled-claude-code-fpf-strategize

What this skill IS

This is strategic betting, not variant selection. You are choosing which class of approach to bet on — scoping the search space for variant generation. "We'll explore variants within [family X] because [reason]."

Strategizing ≠ SoTA survey. SoTA maps the landscape; strategizing bets on a region. Strategizing ≠ variant selection. Strategy scopes the family; selection picks from generated variants.

Prerequisites

  • PROB-* or ANOM-* exists with acceptance spec

  • SOTA-* exists (if missing: invoke /fpf-sota first)

Output

.fpf/decisions/STRAT-${CLAUDE_SESSION_ID}--<slug>.md

Constraints (quality bar)

  • C1: ≥2 candidate method families evaluated

  • C2: Each family assessed for admissibility against PROB-* constraints

  • C3: Primary bet stated with rationale (why this family over others)

  • C4: Variant generation axes defined (what will differ between variants within this family)

  • C5: Invalidation conditions stated (what would cause a pivot to different family)

  • C6: Links to PROB-* and SOTA-*

  • C7: Scaling-Law Lens (SLL): for each family, state how it scales with more resources (compute/data/time). Minimum two measurement points (Scale-Probe) to claim a slope. Elasticity class: growing | plateau | declining. At comparable budgets, prefer the family with better scaling slopes (BLP — Bitter Lesson Preference).

Format

Strategy Card

  • ID: STRAT-... Problem: PROB-... SoTA: SOTA-...
  • valid_until: YYYY-MM-DD

Candidate families

| # | Family | Admissible? | Scaling (SLL) | Key risks | | F1 | ... | Y/N | growing/plateau/declining | ... | | F2 | ... | Y/N | growing/plateau/declining | ... |

Bet

Primary: [family]. Rationale: ...

Variant generation axes

(diversity dimensions within chosen family)

Invalidation conditions

(what would cause a pivot)

valid_until

(when does this strategy bet go stale? Tech landscape changes, new SoTA emerges)

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Coding

fpf-review

No summary provided by upstream source.

Repository SourceNeeds Review
Coding

fpf-problem-portfolio

No summary provided by upstream source.

Repository SourceNeeds Review
Coding

fpf-problem-framing

No summary provided by upstream source.

Repository SourceNeeds Review