Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Survey the codebase for analogous features, reusable utilities, and existing patterns relevant to a proposed change. Returns structured findings without writing code. Use when the user asks to "survey patterns", "find existing patterns", "look for analogous features", "check how similar things are done", "find prior art for this change", or needs pattern context before planning a change.
.claude/skills/tobihagemann-survey-patterns/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -50% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 220% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -51% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 2% | 0% |
Search the codebase for analogous features and reusable building blocks before planning a change, and read any source the task names outside this repo. Returns structured findings. Does not write code or plans.
Determine what the change is about:
State the task back in one sentence to confirm scope before searching.
Spawn a single subagent (model: "opus", no name). Wait for it to report before continuing; do not relaunch it if it has not yet reported. Its prompt must direct it to treat the shared working tree and its git index as read-only and to survey by reading and reasoning. HEAD stays where it is: read other refs with git show <ref>:<path> rather than git checkout or git switch. The subagent's prompt must include:
The subagent covers all three categories (Analogous Features, Reusable Utilities, Convention Anchors) in one sweep, reads any external source the task names, and returns a single structured report.
Output the subagent's report verbatim. Do not reformat or re-synthesize — references/pattern-surveyor.md specifies the exact output format the subagent produces.
Then use the TaskList tool and proceed to any remaining task.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→fail | 10,909 | 26,769 | +145% | 1 | 1 | 0% | 1,423 | 1,124 | -21% | 0 | 0 | — |
case-01 | fail→fail | 15,009 | 30,652 | +104% | 1 | 1 | 0% | 1,442 | 1,219 | -15% | 0 | 0 | — |
case-03 | fail→fail | 14,725 | 5,610 | -62% | 1 | 1 | 0% | 2,391 | 1,115 | -53% | 0 | 0 | — |
case-04 | fail→pass | 13,601 | 5,736 | -58% | 1 | 1 | 0% | 2,447 | 1,228 | -50% | 0 | 0 | — |
case-05 | fail→fail | 8,523 | 10,727 | +26% | 1 | 1 | 0% | 963 | 2,458 | +155% | 0 | 0 | — |
case-06 | fail→pass | 8,848 | 18,481 | +109% | 1 | 1 | 0% | 1,310 | 4,195 | +220% | 0 | 0 | — |
case-07 | fail→fail | 36,147 | 11,485 | -68% | 1 | 1 | 0% | 247 | 988 | +300% | 0 | 0 | — |
case-08 | fail→fail | 12,948 | 6,021 | -53% | 1 | 1 | 0% | 2,630 | 1,299 | -51% | 0 | 0 | — |
case-09 | fail→fail | 15,649 | 16,652 | +6% | 1 | 1 | 0% | 2,838 | 3,662 | +29% | 0 | 0 | — |
case-10 | fail→pass | 15,451 | 5,566 | -64% | 1 | 1 | 0% | 2,721 | 1,327 | -51% | 0 | 0 | — |
case-11 | fail→fail | 10,148 | 4,823 | -52% | 1 | 1 | 0% | 1,499 | 1,042 | -30% | 0 | 0 | — |
case-12 | fail→fail | 8,777 | 5,904 | -33% | 1 | 1 | 0% | 443 | 1,392 | +214% | 0 | 0 | — |
case-13 | fail→pass | 21,298 | 5,156 | -76% | 1 | 1 | 0% | 1,422 | 1,264 | -11% | 0 | 0 | — |
case-14 | fail→pass | 11,397 | 6,411 | -44% | 1 | 1 | 0% | 1,441 | 1,477 | +2% | 0 | 0 | — |
case-20 | fail→pass | 5,656 | 5,854 | +4% | 1 | 1 | 0% | 284 | 1,239 | +336% | 0 | 0 | — |
case-15 | pass→pass | 15,360 | 6,921 | -55% | 1 | 1 | 0% | 3,061 | 1,553 | -49% | 0 | 0 | — |
case-16 | fail→fail | 12,863 | 4,835 | -62% | 1 | 1 | 0% | 2,087 | 903 | -57% | 0 | 0 | — |
case-17 | fail→pass | 12,380 | 5,222 | -58% | 1 | 1 | 0% | 375 | 1,164 | +210% | 0 | 0 | — |
case-18 | fail→fail | 5,302 | 6,640 | +25% | 1 | 1 | 0% | 249 | 1,298 | +421% | 0 | 0 | — |
case-19 | fail→pass | 10,330 | 6,779 | -34% | 1 | 1 | 0% | 1,415 | 1,408 | -0% | 0 | 0 | — |
case-21 | fail→fail | 9,416 | 4,952 | -47% | 1 | 1 | 0% | 726 | 1,033 | +42% | 0 | 0 | — |
case-22 | fail→fail | 24,970 | 26,836 | +7% | 1 | 1 | 0% | 4,333 | 3,299 | -24% | 0 | 0 | — |
case-23 | pass→fail | 9,758 | 6,217 | -36% | 1 | 1 | 0% | 2,138 | 1,418 | -34% | 0 | 0 | — |
case-24 | pass→fail | 9,754 | 6,813 | -30% | 1 | 1 | 0% | 1,904 | 1,502 | -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. 24 cases were attempted, and 18 counted toward the lift figure. The other 6 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 +25 percentage points is the difference between those two pass rates over the 18 comparable cases. 3 cases got worse with the skill loaded, and they are 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/21/2026 | -100% |
Other measured skills in the registry, with their headline benchmark lift.