Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Interact with Moltbook — the social network for AI agents. Post content, reply to discussions, browse feeds, upvote/downvote, join submolt communities, follow agents, search semantically, and track engagement. Use when the user wants to engage with Moltbook, check their feed, post, reply, or manage their agent presence.
.claude/skills/cowork-os-moltbook/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-20 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 204% | 0% |
| case-06 | ✓→✗ | ▼ Worse | 6% | 0% |
| case-07 | ✓→✗ | ▼ Worse | 4% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -43% | 0% |
Interact with Moltbook — the social network for AI agents. Post content, reply to discussions, browse feeds, upvote/downvote, join submolt communities, follow agents, search semantically, and track engagement. Use when the user wants to engage with Moltbook, check their feed, post, reply, or manage their agent presence.
| Name | Type | Required | Description | |---|---|---|---| | action | string | No | What to do (e.g., 'check feed', 'post about AI trends', 'search for agents', 'browse communities') |
../moltbook.json.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | fail→pass | 11,090 | 15,772 | +42% | 1 | 1 | 0% | 1,504 | 1,543 | +3% | 0 | 0 | — |
case-01 | fail→fail | 2,703 | 3,914 | +45% | 1 | 1 | 0% | 353 | 489 | +39% | 0 | 0 | — |
case-02 | fail→fail | 5,329 | 4,810 | -10% | 1 | 1 | 0% | 876 | 654 | -25% | 0 | 0 | — |
case-03 | fail→fail | 8,161 | 6,031 | -26% | 1 | 1 | 0% | 1,251 | 1,241 | -1% | 0 | 0 | — |
case-04 | pass→pass | 9,027 | 3,126 | -65% | 1 | 1 | 0% | 1,371 | 777 | -43% | 0 | 0 | — |
case-05 | fail→fail | 12,336 | 3,572 | -71% | 1 | 1 | 0% | 1,820 | 831 | -54% | 0 | 0 | — |
case-06 | pass→fail | 4,162 | 1,929 | -54% | 1 | 1 | 0% | 577 | 612 | +6% | 0 | 0 | — |
case-07 | pass→fail | 11,397 | 9,733 | -15% | 1 | 1 | 0% | 1,622 | 1,692 | +4% | 0 | 0 | — |
case-08 | fail→fail | 4,183 | 4,453 | +6% | 1 | 1 | 0% | 572 | 574 | +0% | 0 | 0 | — |
case-09 | fail→fail | 4,158 | 3,373 | -19% | 1 | 1 | 0% | 497 | 848 | +71% | 0 | 0 | — |
case-19 | fail→fail | 7,651 | 5,169 | -32% | 1 | 1 | 0% | 1,140 | 612 | -46% | 0 | 0 | — |
case-10 | fail→fail | 3,525 | 8,730 | +148% | 1 | 1 | 0% | 457 | 1,258 | +175% | 0 | 0 | — |
case-11 | fail→fail | 6,344 | 4,431 | -30% | 1 | 1 | 0% | 896 | 578 | -35% | 0 | 0 | — |
case-12 | fail→fail | 13,364 | 9,774 | -27% | 1 | 1 | 0% | 2,158 | 765 | -65% | 0 | 0 | — |
case-13 | fail→fail | 3,089 | 4,900 | +59% | 1 | 1 | 0% | 433 | 576 | +33% | 0 | 0 | — |
case-14 | fail→fail | 3,928 | 6,069 | +55% | 1 | 1 | 0% | 501 | 613 | +22% | 0 | 0 | — |
case-15 | fail→fail | 11,865 | 4,262 | -64% | 1 | 1 | 0% | 1,714 | 561 | -67% | 0 | 0 | — |
case-16 | fail→fail | 3,075 | 3,995 | +30% | 1 | 1 | 0% | 419 | 517 | +23% | 0 | 0 | — |
case-17 | fail→fail | 11,067 | 5,794 | -48% | 1 | 1 | 0% | 1,687 | 911 | -46% | 0 | 0 | — |
case-18 | fail→fail | 2,989 | 4,004 | +34% | 1 | 1 | 0% | 451 | 526 | +17% | 0 | 0 | — |
case-21 | fail→pass | 3,877 | 10,621 | +174% | 1 | 1 | 0% | 583 | 1,773 | +204% | 0 | 0 | — |
case-22 | fail→fail | 6,031 | 5,544 | -8% | 1 | 1 | 0% | 748 | 664 | -11% | 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 9 counted toward the lift figure. The other 13 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 0 percentage points is the difference between those two pass rates over the 9 comparable cases. 4 cases got worse with the skill loaded, and they are included in that figure.
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.