Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Write internal communications such as status reports, leadership updates, 3P updates, newsletters, FAQs, incident reports, and project updates using repeatable internal formats.
.claude/skills/sickn33-internal-comms/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -60% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -56% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -56% | 0% |
Primary editorial path for this compatibility group. The full instructions and support files remain local so existing installations continue to work offline. This is one shared procedure, not an additional capability. Preserve the callable ID when an existing manifest or client configuration uses it. Modified in AAS on 2026-09-05; original metadata and license notices are retained.
To write internal communications, use this skill for:
To write any internal communication:
examples/ directory:examples/3p-updates.md - For Progress/Plans/Problems team updatesexamples/company-newsletter.md - For company-wide newslettersexamples/faq-answers.md - For answering frequently asked questionsexamples/general-comms.md - For anything else that doesn't explicitly match one of the aboveFor other formats, use the provided audience/purpose and the general guide. Ask only for missing information that materially changes the draft.
Use the supplied audience, date range and approved sources. Read only relevant accessible material; reactions, executive seniority and document views are not proof of accuracy. A private source does not automatically belong in a company-wide update. Drafting does not authorize sending, scheduling or publishing the communication.
For a weekly 3P update, turn two confirmed shipped items, one next-week task and one open dependency into Progress/Plans/Problems. Preserve dates and uncertainty, link the source where appropriate and leave unknown metrics out. Expected handoff is a draft for the intended audience, not a claim that a message was sent.
3P updates, company newsletter, company comms, weekly update, faqs, common questions, updates, internal comms
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 12,088 | 9,909 | -18% | 1 | 1 | 0% | 2,042 | 1,918 | -6% | 0 | 0 | — |
case-02 | fail→pass | 8,041 | 1,881 | -77% | 1 | 1 | 0% | 1,213 | 622 | -49% | 0 | 0 | — |
case-03 | fail→pass | 9,495 | 1,841 | -81% | 1 | 1 | 0% | 1,620 | 640 | -60% | 0 | 0 | — |
case-04 | fail→pass | 8,574 | 2,374 | -72% | 1 | 1 | 0% | 1,481 | 645 | -56% | 0 | 0 | — |
case-05 | fail→pass | 8,973 | 2,503 | -72% | 1 | 1 | 0% | 1,743 | 774 | -56% | 0 | 0 | — |
case-06 | fail→pass | 7,694 | 2,239 | -71% | 1 | 1 | 0% | 1,282 | 716 | -44% | 0 | 0 | — |
case-07 | fail→fail | 9,530 | 8,221 | -14% | 1 | 1 | 0% | 1,581 | 1,711 | +8% | 0 | 0 | — |
case-08 | fail→pass | 10,963 | 4,580 | -58% | 1 | 1 | 0% | 1,740 | 1,039 | -40% | 0 | 0 | — |
case-09 | pass→pass | 4,014 | 2,697 | -33% | 1 | 1 | 0% | 736 | 861 | +17% | 0 | 0 | — |
case-10 | fail→pass | 9,641 | 4,965 | -49% | 1 | 1 | 0% | 1,585 | 1,064 | -33% | 0 | 0 | — |
case-11 | pass→pass | 7,700 | 5,824 | -24% | 1 | 1 | 0% | 1,250 | 1,182 | -5% | 0 | 0 | — |
case-12 | fail→fail | 4,358 | 4,515 | +4% | 1 | 1 | 0% | 671 | 1,090 | +62% | 0 | 0 | — |
case-13 | fail→fail | 12,050 | 6,821 | -43% | 1 | 1 | 0% | 1,960 | 1,453 | -26% | 0 | 0 | — |
case-14 | pass→pass | 9,689 | 4,438 | -54% | 1 | 1 | 0% | 1,561 | 948 | -39% | 0 | 0 | — |
case-15 | pass→pass | 6,608 | 3,102 | -53% | 1 | 1 | 0% | 1,090 | 823 | -24% | 0 | 0 | — |
case-16 | pass→pass | 9,804 | 5,727 | -42% | 1 | 1 | 0% | 1,529 | 1,142 | -25% | 0 | 0 | — |
case-17 | fail→pass | 9,457 | 2,726 | -71% | 1 | 1 | 0% | 1,444 | 790 | -45% | 0 | 0 | — |
case-18 | pass→fail | 13,017 | 4,010 | -69% | 1 | 1 | 0% | 2,178 | 953 | -56% | 0 | 0 | — |
case-19 | pass→pass | 9,218 | 11,532 | +25% | 1 | 1 | 0% | 1,629 | 2,149 | +32% | 0 | 0 | — |
case-20 | pass→pass | 10,889 | 11,528 | +6% | 1 | 1 | 0% | 1,809 | 2,087 | +15% | 0 | 0 | — |
case-21 | pass→pass | 8,642 | 4,406 | -49% | 1 | 1 | 0% | 1,544 | 1,005 | -35% | 0 | 0 | — |
case-22 | pass→pass | 10,404 | 2,414 | -77% | 1 | 1 | 0% | 1,533 | 707 | -54% | 0 | 0 | — |
case-23 | fail→pass | 11,033 | 6,702 | -39% | 1 | 1 | 0% | 1,764 | 1,228 | -30% | 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. 23 cases were attempted. The headline lift of +39 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
The publisher has shipped newer versions since this run, so these numbers describe v1, not the version currently listed.
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.