Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Trace Elixir call trees from entry points via mix xref. Use when debugging data flow, planning signature changes, or understanding how a bug reaches code.
.claude/skills/oliver-kriska-trace/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -35% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -40% | 0% |
Build call trees showing how functions are reached from entry points.
mix xref callers first - It's authoritative; grep is fallback only| Condition | Why Call Tree Helps | |-----------|---------------------| | Unexpected nil/value at runtime | Trace where the value originates | | Bug can't reproduce locally | See all entry points that reach the code | | Changing function signature | Find all callers and their argument patterns | | Incomplete stack trace | Get full path context | | "Where does X come from?" | Visual answer to data flow question |
Run the caller query first, then inspect another function in the chain as needed:
bashmix xref callers MyApp.Accounts.update_user/2 mix xref callers MyApp.Accounts.get_user/1
Read the reported locations to see argument patterns.
| Pattern | Type | |---------|------| | def mount/3, def handle_event/3 | LiveView | | def index/2, def show/2, def create/2 | Controller | | def perform(%Oban.Job{}) | Oban Worker | | def handle_call/3, def handle_cast/2 | GenServer |
For full recursive tree with argument extraction and parallel category tracing:
Determine the effective maximum nesting depth. Use an explicit positive-integer CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH value first; when it is unset, inspect claude --version (the default is 1 in 2.1.217–2.1.218 and 3 in 2.1.219+). If the version is unavailable, conservatively use 1. At depth 2+, delegate to the orchestrator below. At depth 1, keep orchestration in this main session: spawn the applicable controller, LiveView, worker, and internal tracing prompts directly, then merge their results. Never spawn an orchestrator that cannot delegate.
Agent(subagent_type: "phx:call-tracer", prompt: "Build call tree for MyApp.Accounts.update_user/2")The call-tracer agent uses parallel subagents for each entry point category:
Each gets fresh 200k context for deep exploration.
.claude/plans/{slug}/research/call-tree-{function}.md
For detailed patterns:
${CLAUDE_SKILL_DIR}/references/mix-xref-usage.md - Full mix xref commands and options${CLAUDE_SKILL_DIR}/references/entry-points.md - All Phoenix/OTP entry point patterns${CLAUDE_SKILL_DIR}/references/argument-extraction.md - AST parsing for argument patterns| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 17,176 | 5,089 | -70% | 1 | 1 | 0% | 2,874 | 1,094 | -62% | 0 | 0 | — |
case-02 | fail→fail | 9,839 | 5,964 | -39% | 1 | 1 | 0% | 1,668 | 1,213 | -27% | 0 | 0 | — |
case-03 | fail→fail | 35,767 | 6,372 | -82% | 1 | 1 | 0% | 3,271 | 1,229 | -62% | 0 | 0 | — |
case-04 | pass→pass | 13,944 | 5,939 | -57% | 1 | 1 | 0% | 2,240 | 1,789 | -20% | 0 | 0 | — |
case-05 | pass→pass | 10,998 | 3,494 | -68% | 1 | 1 | 0% | 1,769 | 1,363 | -23% | 0 | 0 | — |
case-10 | pass→pass | 12,192 | 2,419 | -80% | 1 | 1 | 0% | 1,874 | 1,218 | -35% | 0 | 0 | — |
case-06 | pass→pass | 12,056 | 2,663 | -78% | 1 | 1 | 0% | 2,000 | 1,231 | -38% | 0 | 0 | — |
case-07 | fail→pass | 11,621 | 2,763 | -76% | 1 | 1 | 0% | 1,925 | 1,244 | -35% | 0 | 0 | — |
case-08 | fail→pass | 12,049 | 4,393 | -64% | 1 | 1 | 0% | 1,877 | 1,481 | -21% | 0 | 0 | — |
case-09 | pass→pass | 12,745 | 8,808 | -31% | 1 | 1 | 0% | 2,127 | 2,198 | +3% | 0 | 0 | — |
case-11 | fail→fail | 7,464 | 2,300 | -69% | 1 | 1 | 0% | 1,276 | 1,213 | -5% | 0 | 0 | — |
case-12 | pass→pass | 13,194 | 3,030 | -77% | 1 | 1 | 0% | 2,214 | 1,328 | -40% | 0 | 0 | — |
case-13 | pass→pass | 15,434 | 5,258 | -66% | 1 | 1 | 0% | 2,765 | 1,834 | -34% | 0 | 0 | — |
case-14 | pass→pass | 14,708 | 5,743 | -61% | 1 | 1 | 0% | 2,330 | 1,729 | -26% | 0 | 0 | — |
case-15 | pass→pass | 4,115 | 1,897 | -54% | 1 | 1 | 0% | 707 | 1,111 | +57% | 0 | 0 | — |
case-16 | fail→pass | 14,739 | 2,323 | -84% | 1 | 1 | 0% | 1,294 | 1,285 | -1% | 0 | 0 | — |
case-17 | fail→pass | 6,640 | 2,405 | -64% | 1 | 1 | 0% | 1,215 | 1,225 | +1% | 0 | 0 | — |
case-18 | fail→pass | 10,557 | 1,799 | -83% | 1 | 1 | 0% | 1,710 | 1,033 | -40% | 0 | 0 | — |
case-19 | pass→pass | 8,356 | 3,244 | -61% | 1 | 1 | 0% | 1,354 | 1,303 | -4% | 0 | 0 | — |
case-20 | fail→fail | 12,164 | 6,644 | -45% | 1 | 1 | 0% | 2,254 | 2,079 | -8% | 0 | 0 | — |
case-21 | pass→pass | 11,408 | 7,355 | -36% | 1 | 1 | 0% | 2,156 | 2,027 | -6% | 0 | 0 | — |
case-22 | pass→pass | 9,050 | 7,609 | -16% | 1 | 1 | 0% | 1,658 | 2,109 | +27% | 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 19 counted toward the lift figure. The other 3 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 +23 percentage points is the difference between those two pass rates over the 19 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.