Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Reset the FPF reasoning cycle to start fresh
.claude/skills/bilal140202-reset/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 184% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 17% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 31% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -13% | 0% |
Reset the FPF reasoning cycle to start fresh.
Create a session archive and clear active work:
Create a file in .fpf/sessions/ to record the completed/abandoned session:
markdown# In .fpf/sessions/session-2025-01-15-reset.md --- id: session-2025-01-15-reset action: reset created: 2025-01-15T16:00:00Z reason: user_requested --- # Session Archive: 2025-01-15 **Reset Reason**: User requested fresh start ## State at Reset ### Hypotheses - L0: 2 (proposed) - L1: 1 (verified) - L2: 0 (validated) - Invalid: 1 (rejected) ### Files Archived - .fpf/knowledge/L0/hypothesis-a.md - .fpf/knowledge/L0/hypothesis-b.md - .fpf/knowledge/L1/hypothesis-c.md ### Decision Status No decision was finalized. ## Notes Session ended without decision. Hypotheses preserved for potential future reference.
If user wants to clear the knowledge directories:
bashmkdir -p .fpf/archive/session-2025-01-15 mv .fpf/knowledge/L0/*.md .fpf/archive/session-2025-01-15/ 2>/dev/null || true mv .fpf/knowledge/L1/*.md .fpf/archive/session-2025-01-15/ 2>/dev/null || true mv .fpf/knowledge/L2/*.md .fpf/archive/session-2025-01-15/ 2>/dev/null || true
markdown## Reset Complete Session archived to: .fpf/sessions/session-2025-01-15-reset.md Current state: - L0: 0 hypotheses - L1: 0 hypotheses - L2: 0 hypotheses Ready for new reasoning cycle. Run `/fpf:propose-hypotheses` to start.
WARNING: This permanently deletes all FPF data.
bashrm -rf .fpf/knowledge/L0/*.md rm -rf .fpf/knowledge/L1/*.md rm -rf .fpf/knowledge/L2/*.md rm -rf .fpf/knowledge/invalid/*.md rm -rf .fpf/evidence/*.md rm -rf .fpf/decisions/*.md
Only do this if explicitly requested by user.
If user wants to re-evaluate existing hypotheses:
bash# Re-audit: L2 -> L1 mv .fpf/knowledge/L2/*.md .fpf/knowledge/L1/ # Re-verify: L1 -> L0 mv .fpf/knowledge/L1/*.md .fpf/knowledge/L0/
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 8,976 | 4,989 | -44% | 1 | 1 | 0% | 1,215 | 1,054 | -13% | 0 | 0 | — |
case-11 | fail→pass | 8,285 | 2,808 | -66% | 1 | 1 | 0% | 1,353 | 1,237 | -9% | 0 | 0 | — |
case-02 | fail→pass | 3,166 | 2,323 | -27% | 1 | 1 | 0% | 443 | 1,256 | +184% | 0 | 0 | — |
case-03 | fail→fail | 7,857 | 3,666 | -53% | 1 | 1 | 0% | 1,137 | 1,150 | +1% | 0 | 0 | — |
case-04 | fail→pass | 6,583 | 3,326 | -49% | 1 | 1 | 0% | 1,242 | 1,448 | +17% | 0 | 0 | — |
case-05 | fail→pass | 19,356 | 4,695 | -76% | 1 | 1 | 0% | 1,336 | 1,744 | +31% | 0 | 0 | — |
case-06 | fail→pass | 8,636 | 2,137 | -75% | 1 | 1 | 0% | 1,342 | 1,169 | -13% | 0 | 0 | — |
case-07 | fail→pass | 14,372 | 4,621 | -68% | 1 | 1 | 0% | 2,392 | 1,512 | -37% | 0 | 0 | — |
case-08 | fail→pass | 12,567 | 3,350 | -73% | 1 | 1 | 0% | 1,604 | 1,343 | -16% | 0 | 0 | — |
case-09 | fail→pass | 7,057 | 2,557 | -64% | 1 | 1 | 0% | 1,303 | 1,213 | -7% | 0 | 0 | — |
case-10 | fail→pass | 7,857 | 2,509 | -68% | 1 | 1 | 0% | 1,176 | 1,116 | -5% | 0 | 0 | — |
case-12 | fail→pass | 14,358 | 4,546 | -68% | 1 | 1 | 0% | 2,568 | 1,679 | -35% | 0 | 0 | — |
case-13 | fail→fail | 11,323 | 4,871 | -57% | 1 | 1 | 0% | 1,718 | 1,671 | -3% | 0 | 0 | — |
case-14 | fail→pass | 7,019 | 3,072 | -56% | 1 | 1 | 0% | 1,023 | 1,289 | +26% | 0 | 0 | — |
case-15 | pass→pass | 7,682 | 3,922 | -49% | 1 | 1 | 0% | 1,115 | 1,546 | +39% | 0 | 0 | — |
case-16 | fail→pass | 9,987 | 2,959 | -70% | 1 | 1 | 0% | 1,587 | 1,354 | -15% | 0 | 0 | — |
case-17 | fail→pass | 11,539 | 7,836 | -32% | 1 | 1 | 0% | 1,651 | 2,122 | +29% | 0 | 0 | — |
case-18 | pass→pass | 12,529 | 7,679 | -39% | 1 | 1 | 0% | 2,267 | 2,216 | -2% | 0 | 0 | — |
case-19 | fail→pass | 9,478 | 3,697 | -61% | 1 | 1 | 0% | 1,468 | 1,472 | +0% | 0 | 0 | — |
case-20 | pass→pass | 6,308 | 4,302 | -32% | 1 | 1 | 0% | 966 | 1,468 | +52% | 0 | 0 | — |
case-21 | fail→pass | 11,795 | 4,631 | -61% | 1 | 1 | 0% | 1,756 | 1,666 | -5% | 0 | 0 | — |
case-22 | pass→pass | 7,995 | 4,891 | -39% | 1 | 1 | 0% | 1,136 | 1,608 | +42% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 20 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +68 percentage points is the difference between those two pass rates over the 20 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.