Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use Hermes Tweet and Xquik for X/Twitter agent workflows. Plan social listening, account and follower analysis, post research, monitors, webhook alerts, REST API calls, MCP client setup, and safe tweet actions.
.claude/skills/microck-hermes-tweet-xquik/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -51% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -23% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -23% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -43% | 0% |
Use this skill when an agent needs X/Twitter workflows through Hermes Tweet or Xquik.
https://github.com/Xquik-dev/hermes-tweethttps://docs.xquik.com/llms.txthttps://xquik.com/openapi.jsonhttps://xquik.com/.well-known/mcp.jsonhttps://xquik.com/mcphttps://github.com/Xquik-dev/hermes-tweet when the runtime is Hermes Agent.XQUIK_API_KEY in the agent or MCP client secret store for authenticated read and action workflows.HERMES_TWEET_ENABLE_ACTIONS=false (disabled) unless the user explicitly approved tweet actions.XQUIK_API_KEY and HERMES_TWEET_ENABLE_ACTIONS=true.bashcurl -fsSL https://xquik.com/.well-known/mcp.json | jq . curl -fsSL https://xquik.com/openapi.json | jq '.info.title'
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 20,323 | 7,055 | -65% | 1 | 1 | 0% | 3,551 | 1,742 | -51% | 0 | 0 | — |
case-02 | fail→pass | 17,173 | 8,917 | -48% | 1 | 1 | 0% | 2,742 | 2,094 | -24% | 0 | 0 | — |
case-03 | fail→pass | 18,338 | 10,847 | -41% | 1 | 1 | 0% | 2,930 | 2,245 | -23% | 0 | 0 | — |
case-04 | pass→pass | 15,402 | 13,311 | -14% | 1 | 1 | 0% | 2,912 | 3,008 | +3% | 0 | 0 | — |
case-05 | pass→pass | 14,555 | 11,670 | -20% | 1 | 1 | 0% | 2,735 | 2,738 | +0% | 0 | 0 | — |
case-10 | fail→pass | 6,119 | 1,647 | -73% | 1 | 1 | 0% | 955 | 738 | -23% | 0 | 0 | — |
case-06 | pass→pass | 12,924 | 7,240 | -44% | 1 | 1 | 0% | 2,295 | 1,766 | -23% | 0 | 0 | — |
case-07 | pass→pass | 8,267 | 2,090 | -75% | 1 | 1 | 0% | 1,413 | 757 | -46% | 0 | 0 | — |
case-08 | fail→pass | 7,983 | 2,174 | -73% | 1 | 1 | 0% | 1,445 | 824 | -43% | 0 | 0 | — |
case-09 | pass→pass | 11,790 | 3,675 | -69% | 1 | 1 | 0% | 1,901 | 920 | -52% | 0 | 0 | — |
case-11 | pass→pass | 11,182 | 2,919 | -74% | 1 | 1 | 0% | 1,744 | 946 | -46% | 0 | 0 | — |
case-12 | pass→pass | 12,289 | 5,188 | -58% | 1 | 1 | 0% | 2,078 | 1,373 | -34% | 0 | 0 | — |
case-13 | fail→pass | 5,027 | 1,910 | -62% | 1 | 1 | 0% | 704 | 751 | +7% | 0 | 0 | — |
case-14 | pass→pass | 12,177 | 4,359 | -64% | 1 | 1 | 0% | 1,779 | 1,040 | -42% | 0 | 0 | — |
case-15 | fail→pass | 7,085 | 1,861 | -74% | 1 | 1 | 0% | 1,027 | 690 | -33% | 0 | 0 | — |
case-16 | fail→pass | 9,605 | 2,277 | -76% | 1 | 1 | 0% | 1,392 | 873 | -37% | 0 | 0 | — |
case-17 | pass→pass | 5,599 | 1,744 | -69% | 1 | 1 | 0% | 880 | 678 | -23% | 0 | 0 | — |
case-18 | fail→pass | 8,874 | 1,135 | -87% | 1 | 1 | 0% | 1,461 | 632 | -57% | 0 | 0 | — |
case-19 | pass→pass | 9,496 | 1,818 | -81% | 1 | 1 | 0% | 1,633 | 732 | -55% | 0 | 0 | — |
case-20 | pass→pass | 6,590 | 2,037 | -69% | 1 | 1 | 0% | 876 | 735 | -16% | 0 | 0 | — |
case-21 | pass→pass | 17,761 | 9,750 | -45% | 1 | 1 | 0% | 2,758 | 1,961 | -29% | 0 | 0 | — |
case-22 | fail→pass | 12,415 | 2,592 | -79% | 1 | 1 | 0% | 1,990 | 840 | -58% | 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. The headline lift of +45 percentage points is the difference between those two pass rates over the 22 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.