ReddRadar
Agent skill
call-trace
Trace the call path between two functions by following callers/callees. Use when the user asks how function A reaches function B, wants to see connections between symbols, or asks "what calls what."
Install this agent skill to your Project
npx add-skill https://github.com/anortham/julie/tree/main/.claude/skills/call-trace
SKILL.md
Call Trace
Trace the call path from a starting function to an ending function.
Process
Step 0: Resolve Symbol Names
If the symbol names are approximate or ambiguous, resolve them first:
fast_search(query="<name>", search_target="definitions")
If deep_dive returns the wrong symbol (e.g., a common name like new or process), use the context_file parameter to disambiguate:
deep_dive(symbol="process", context_file="handler")
Step 1: Orient (Optional)
If both symbols are in the same domain, get a broad view first:
get_context(query="<shared concept or module name>")
This may reveal the connection immediately from the pivots and neighbors.
Step 2: Build Forward Call Graph (Start → End)
Starting from the start function, follow callees:
deep_dive(symbol="<start>", depth="context")— get callees list- For each callee:
- If callee matches end function → path found
- If callee already visited → skip (cycle)
- If depth > 5 → stop (too deep)
- Otherwise,
deep_dive(symbol="<callee>", depth="overview")and recurse
Track the path as you go: [start, callee_1, callee_2, ..., end]
Step 3: Try Reverse If Forward Fails
If forward search doesn't find a path, try backward from the end function:
fast_refs(symbol="<end>", reference_kind="call")— get callers- For each caller, check if it's reachable from start
Step 4: Report Results
Path found:
Call path: start_fn → helper_fn → process_fn → end_fn (3 steps)
Details:
1. start_fn (src/main.rs:45) calls helper_fn
2. helper_fn (src/utils.rs:120) calls process_fn
3. process_fn (src/handler.rs:78) calls end_fn (src/tools.rs:30)
No path found:
No call path found between start_fn and end_fn within 5 steps.
start_fn callees: [a, b, c]
end_fn callers: [x, y, z]
These don't overlap — the functions may not be connected.
Important Notes
- Max depth: 5 — deeper traces are usually noise
- Track visited set — avoid infinite loops from recursive calls
- Prefer shortest path — if multiple paths exist, the first found (BFS-like) is usually most direct
- Generic names (e.g.,
new,from,into) should be filtered — they create false connections - Reference workspaces: Pass
workspace: "<workspace_id>"to all tool calls when tracing in a non-primary workspace
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
anortham/julie
logic-flow
Explain a function's logic step-by-step by analyzing its implementation and call graph. Use when the user asks "how does this work", "walk me through this", or wants to understand control flow and decision points.
anortham/julie
metrics
Show Julie operational metrics -- session stats, tool usage, context efficiency, and historical trends. Use when the user asks about Julie's performance, how it's doing, how much context was saved, or wants a metrics/stats report.
anortham/julie
architecture
Generate an architecture overview — key entry points, module map, dependency flow, and suggested reading order. Use when the user is new to a codebase, asks "how does this work?", wants an architecture overview, or needs onboarding documentation.
anortham/julie
dependency-graph
Show module dependencies by analyzing imports, exports, and cross-references. Use when the user asks what a file imports, what depends on a module, or wants to see dependency structure.
anortham/julie
search-debug
Diagnose why a search returns unexpected results — analyze scoring factors, tokenization, and index health for dogfooding Julie's search quality. Use when a Julie search misses an expected symbol, returns wrong results, ranks something surprisingly low, or when investigating search quality issues during development.
anortham/julie
impact-analysis
Didn't find tool you were looking for?