Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Expert-level agent skills for the entire ServiceNow development ecosystem. Covers Fluent SDK metadata (.now.ts), legacy scripting (GlideRecord, Script Includes, Client Scripts, g_form), and CLI setup (now-sdk auth, build, install). Use when generating, reviewing, or explaining any ServiceNow code.
.claude/skills/bilal140202-servicenow-agent-skills/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -13% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-24 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-07 | ✓→✗ | ▼ Worse | -50% | 0% |
This is a multi-skill suite. Each sub-skill contains its own SKILL.md with hard rules, reference files, and example assets. Do not pre-load all sub-skills. Read only the one that matches the current task.
| Task Signal | Sub-Skill to Read | |-------------|-------------------| | .now.ts files, Fluent SDK imports, @servicenow/sdk/core, Table/ACL/BusinessRule metadata, now.config.json | ./sn-sdk-fluent/SKILL.md | | GlideRecord, Script Includes, Business Rules (legacy), Client Scripts, g_form, g_user, gs.log | ./sn-scripting/SKILL.md | | now-sdk CLI commands, project scaffolding, OAuth/Basic auth setup, environment configuration | ./sn-sdk-setup/SKILL.md |
SKILL.md before writing any code.now.config.json or .now.ts files, the project uses the Fluent SDK — do not generate GlideRecord code. If the workspace has no Fluent markers, use sn-scripting.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-10 | pass→pass | 4,350 | 2,970 | -32% | 1 | 1 | 0% | 835 | 934 | +12% | 0 | 0 | — |
case-01 | fail→pass | 11,135 | 10,784 | -3% | 1 | 1 | 0% | 2,317 | 2,010 | -13% | 0 | 0 | — |
case-02 | fail→fail | 10,840 | 3,039 | -72% | 1 | 1 | 0% | 2,155 | 807 | -63% | 0 | 0 | — |
case-03 | pass→pass | 3,621 | 3,231 | -11% | 1 | 1 | 0% | 711 | 1,071 | +51% | 0 | 0 | — |
case-04 | fail→fail | 11,777 | 7,473 | -37% | 1 | 1 | 0% | 2,291 | 1,796 | -22% | 0 | 0 | — |
case-05 | fail→fail | 7,319 | 8,573 | +17% | 1 | 1 | 0% | 1,445 | 1,999 | +38% | 0 | 0 | — |
case-06 | pass→pass | 9,259 | 5,747 | -38% | 1 | 1 | 0% | 1,890 | 1,568 | -17% | 0 | 0 | — |
case-07 | pass→fail | 5,341 | 3,291 | -38% | 1 | 1 | 0% | 1,105 | 551 | -50% | 0 | 0 | — |
case-08 | pass→pass | 9,220 | 6,990 | -24% | 1 | 1 | 0% | 1,692 | 1,716 | +1% | 0 | 0 | — |
case-09 | pass→pass | 6,978 | 9,306 | +33% | 1 | 1 | 0% | 1,344 | 2,306 | +72% | 0 | 0 | — |
case-11 | pass→pass | 7,430 | 3,152 | -58% | 1 | 1 | 0% | 1,308 | 902 | -31% | 0 | 0 | — |
case-12 | fail→fail | 4,259 | 2,611 | -39% | 1 | 1 | 0% | 746 | 835 | +12% | 0 | 0 | — |
case-13 | pass→pass | 4,569 | 6,211 | +36% | 1 | 1 | 0% | 856 | 1,261 | +47% | 0 | 0 | — |
case-14 | fail→pass | 8,614 | 3,546 | -59% | 1 | 1 | 0% | 1,594 | 918 | -42% | 0 | 0 | — |
case-15 | fail→pass | 11,816 | 5,440 | -54% | 1 | 1 | 0% | 2,443 | 1,413 | -42% | 0 | 0 | — |
case-16 | fail→fail | 10,329 | 2,092 | -80% | 1 | 1 | 0% | 1,963 | 598 | -70% | 0 | 0 | — |
case-17 | fail→fail | 12,292 | 12,439 | +1% | 1 | 1 | 0% | 2,492 | 2,341 | -6% | 0 | 0 | — |
case-18 | fail→fail | 8,457 | 5,731 | -32% | 1 | 1 | 0% | 1,738 | 1,464 | -16% | 0 | 0 | — |
case-19 | pass→pass | 8,272 | 5,376 | -35% | 1 | 1 | 0% | 1,579 | 1,345 | -15% | 0 | 0 | — |
case-20 | pass→pass | 12,486 | 12,894 | +3% | 1 | 1 | 0% | 2,653 | 3,411 | +29% | 0 | 0 | — |
case-21 | pass→pass | 7,970 | 7,262 | -9% | 1 | 1 | 0% | 1,592 | 1,775 | +11% | 0 | 0 | — |
case-22 | fail→fail | 12,090 | 6,062 | -50% | 1 | 1 | 0% | 2,287 | 1,609 | -30% | 0 | 0 | — |
case-23 | fail→fail | 10,352 | 7,594 | -27% | 1 | 1 | 0% | 1,905 | 1,798 | -6% | 0 | 0 | — |
case-24 | fail→pass | 9,178 | 5,981 | -35% | 1 | 1 | 0% | 1,637 | 1,454 | -11% | 0 | 0 | — |
case-25 | pass→pass | 8,307 | 7,152 | -14% | 1 | 1 | 0% | 1,434 | 1,252 | -13% | 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. 25 cases were attempted, and 24 counted toward the lift figure. The other 1 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 +12 percentage points is the difference between those two pass rates over the 24 comparable cases. 1 case got worse with the skill loaded, and it is 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.