Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Analyze Claude Code sessions via Braintrust
.claude/skills/braintrust-analyze/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -73% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -52% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -56% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -52% | 0% |
Analyze your Claude Code sessions for patterns, issues, and insights using Braintrust tracing data.
Run from the project directory:
bash# Analyze last session - summary with tool/agent/skill breakdown uv run python -m runtime.harness scripts/braintrust_analyze.py --last-session # List recent sessions uv run python -m runtime.harness scripts/braintrust_analyze.py --sessions 5 # Agent usage statistics (last 7 days) uv run python -m runtime.harness scripts/braintrust_analyze.py --agent-stats # Skill usage statistics (last 7 days) uv run python -m runtime.harness scripts/braintrust_analyze.py --skill-stats # Detect loops - find repeated tool patterns (>5 same tool calls) uv run python -m runtime.harness scripts/braintrust_analyze.py --detect-loops # Replay specific session - show full sequence of actions uv run python -m runtime.harness scripts/braintrust_analyze.py --replay <session-id> # Weekly summary - daily activity breakdown uv run python -m runtime.harness scripts/braintrust_analyze.py --weekly-summary # Token trends - usage over time uv run python -m runtime.harness scripts/braintrust_analyze.py --token-trends
--project NAME - Braintrust project name (default: agentica)Find sessions where the same tool was called repeatedly, which may indicate:
bash# What happened in my last session? uv run python -m runtime.harness scripts/braintrust_analyze.py --last-session
Output:
## Session Analysis
**ID:** `92940b91...`
**Started:** 2025-12-24T01:31:05Z
**Spans:** 14
### Tool Usage
- Read: 4
- Bash: 2
- Edit: 2
...bashuv run python -m runtime.harness scripts/braintrust_analyze.py --detect-loops
bashuv run python -m runtime.harness scripts/braintrust_analyze.py --weekly-summary
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 6,429 | 4,868 | -24% | 1 | 1 | 0% | 1,004 | 948 | -6% | 0 | 0 | — |
case-02 | fail→fail | 7,937 | 1,382 | -83% | 1 | 1 | 0% | 1,386 | 908 | -34% | 0 | 0 | — |
case-03 | fail→fail | 5,891 | 5,391 | -8% | 1 | 1 | 0% | 1,044 | 969 | -7% | 0 | 0 | — |
case-04 | fail→pass | 18,905 | 1,748 | -91% | 1 | 1 | 0% | 3,788 | 1,025 | -73% | 0 | 0 | — |
case-13 | fail→pass | 8,944 | 2,239 | -75% | 1 | 1 | 0% | 1,746 | 1,098 | -37% | 0 | 0 | — |
case-05 | fail→pass | 12,589 | 2,014 | -84% | 1 | 1 | 0% | 2,211 | 1,053 | -52% | 0 | 0 | — |
case-06 | fail→pass | 11,871 | 2,614 | -78% | 1 | 1 | 0% | 2,260 | 992 | -56% | 0 | 0 | — |
case-07 | fail→pass | 10,580 | 1,478 | -86% | 1 | 1 | 0% | 1,989 | 954 | -52% | 0 | 0 | — |
case-08 | fail→pass | 14,592 | 2,611 | -82% | 1 | 1 | 0% | 2,311 | 1,167 | -50% | 0 | 0 | — |
case-09 | fail→pass | 3,504 | 1,744 | -50% | 1 | 1 | 0% | 615 | 990 | +61% | 0 | 0 | — |
case-10 | fail→pass | 7,978 | 2,850 | -64% | 1 | 1 | 0% | 1,274 | 1,193 | -6% | 0 | 0 | — |
case-11 | pass→pass | 2,903 | 2,408 | -17% | 1 | 1 | 0% | 459 | 949 | +107% | 0 | 0 | — |
case-12 | fail→pass | 8,471 | 1,373 | -84% | 1 | 1 | 0% | 1,266 | 898 | -29% | 0 | 0 | — |
case-14 | fail→pass | 6,303 | 1,893 | -70% | 1 | 1 | 0% | 1,145 | 1,018 | -11% | 0 | 0 | — |
case-15 | fail→pass | 12,216 | 9,123 | -25% | 1 | 1 | 0% | 1,966 | 2,211 | +12% | 0 | 0 | — |
case-16 | fail→pass | 7,048 | 3,448 | -51% | 1 | 1 | 0% | 1,150 | 1,236 | +7% | 0 | 0 | — |
case-17 | fail→pass | 9,732 | 2,870 | -71% | 1 | 1 | 0% | 1,749 | 1,296 | -26% | 0 | 0 | — |
case-18 | pass→pass | 5,705 | 1,518 | -73% | 1 | 1 | 0% | 968 | 938 | -3% | 0 | 0 | — |
case-19 | fail→pass | 7,611 | 2,237 | -71% | 1 | 1 | 0% | 1,509 | 943 | -38% | 0 | 0 | — |
case-20 | pass→pass | 11,421 | 6,733 | -41% | 1 | 1 | 0% | 2,199 | 1,994 | -9% | 0 | 0 | — |
case-21 | pass→pass | 6,018 | 4,346 | -28% | 1 | 1 | 0% | 1,126 | 1,444 | +28% | 0 | 0 | — |
case-22 | pass→pass | 4,957 | 2,548 | -49% | 1 | 1 | 0% | 932 | 1,124 | +21% | 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 +64 percentage points is the difference between those two pass rates over the 21 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 7/28/2026 | +55% |
Other measured skills in the registry, with their headline benchmark lift.