Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Letta framework for building stateful AI agents with long-term memory. Use for AI agent development, memory management, tool integration, and multi-agent systems.
.claude/skills/majiayu000-letta/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-10 | ✓→✗ | ▼ Worse | 134% | 0% |
| case-03 | ✓→✓ | = Same ✓ | -9% | 0% |
| case-04 | ✓→✓ | = Same ✓ | -20% | 0% |
Comprehensive assistance with letta development, generated from official documentation.
This skill should be triggered when:
Quick reference patterns will be added as you use the skill.
Example 1 (bash):
bashpip install letta-evals
Example 2 (bash):
bashuv pip install letta-evals
Example 3 (python):
pythonimport requests url = "https://api.letta.com/v1/mcp-servers/mcp_server_id" headers = {"Authorization": "Bearer <token>"} response = requests.delete(url, headers=headers) print(response.json())
Example 4 (python):
pythonfrom letta_client import Letta client = Letta( project="YOUR_PROJECT", token="YOUR_TOKEN", ) client.groups.delete( group_id="group-123e4567-e89b-42d3-8456-426614174000", )
This skill includes comprehensive documentation in references/:
Use view to read specific reference files when detailed information is needed.
Start with the getting_started or tutorials reference files for foundational concepts.
Use the appropriate category reference file (api, guides, etc.) for detailed information.
The quick reference section above contains common patterns extracted from the official docs.
Organized documentation extracted from official sources. These files contain:
Add helper scripts here for common automation tasks.
Add templates, boilerplate, or example projects here.
To refresh this skill with updated documentation:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 11,729 | 8,849 | -25% | 1 | 1 | 0% | 1,543 | 1,467 | -5% | 0 | 0 | — |
case-02 | fail→pass | 6,713 | 2,566 | -62% | 1 | 1 | 0% | 1,310 | 1,192 | -9% | 0 | 0 | — |
case-03 | pass→pass | 10,931 | 6,855 | -37% | 1 | 1 | 0% | 1,055 | 965 | -9% | 0 | 0 | — |
case-04 | pass→pass | 14,882 | 8,952 | -40% | 1 | 1 | 0% | 1,840 | 1,464 | -20% | 0 | 0 | — |
case-21 | pass→pass | 9,887 | 10,219 | +3% | 1 | 1 | 0% | 1,840 | 2,592 | +41% | 0 | 0 | — |
case-05 | pass→pass | 5,855 | 11,916 | +104% | 1 | 1 | 0% | 1,033 | 1,909 | +85% | 0 | 0 | — |
case-06 | pass→pass | 7,597 | 5,091 | -33% | 1 | 1 | 0% | 399 | 1,040 | +161% | 0 | 0 | — |
case-07 | pass→pass | 9,166 | 7,927 | -14% | 1 | 1 | 0% | 641 | 975 | +52% | 0 | 0 | — |
case-08 | pass→pass | 9,015 | 3,294 | -63% | 1 | 1 | 0% | 631 | 1,227 | +94% | 0 | 0 | — |
case-22 | pass→pass | 7,976 | 12,470 | +56% | 1 | 1 | 0% | 1,532 | 2,138 | +40% | 0 | 0 | — |
case-09 | pass→pass | 4,542 | 2,477 | -45% | 1 | 1 | 0% | 719 | 1,039 | +45% | 0 | 0 | — |
case-10 | pass→fail | 3,407 | 11,350 | +233% | 1 | 1 | 0% | 446 | 1,042 | +134% | 0 | 0 | — |
case-11 | pass→pass | 4,550 | 1,991 | -56% | 1 | 1 | 0% | 754 | 985 | +31% | 0 | 0 | — |
case-12 | pass→pass | 1,930 | 10,031 | +420% | 1 | 1 | 0% | 293 | 1,464 | +400% | 0 | 0 | — |
case-13 | pass→pass | 15,749 | 6,849 | -57% | 1 | 1 | 0% | 2,094 | 1,900 | -9% | 0 | 0 | — |
case-14 | pass→pass | 12,067 | 2,319 | -81% | 1 | 1 | 0% | 1,080 | 1,039 | -4% | 0 | 0 | — |
case-15 | pass→pass | 13,195 | 2,655 | -80% | 1 | 1 | 0% | 1,361 | 1,067 | -22% | 0 | 0 | — |
case-16 | pass→pass | 6,280 | 2,614 | -58% | 1 | 1 | 0% | 1,068 | 1,112 | +4% | 0 | 0 | — |
case-17 | pass→pass | 10,785 | 10,078 | -7% | 1 | 1 | 0% | 839 | 1,555 | +85% | 0 | 0 | — |
case-18 | pass→pass | 7,114 | 4,400 | -38% | 1 | 1 | 0% | 1,181 | 1,465 | +24% | 0 | 0 | — |
case-19 | pass→pass | 3,175 | 8,778 | +176% | 1 | 1 | 0% | 521 | 1,294 | +148% | 0 | 0 | — |
case-20 | pass→pass | 12,630 | 10,041 | -20% | 1 | 1 | 0% | 1,448 | 1,640 | +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. 22 cases were attempted, and 21 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 +5 percentage points is the difference between those two pass rates over the 21 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.