BenchmarkTracked since May 19, 2026

Symbol-query results in Serena still trigger same-file follow-up reads

A 21-day telemetry thread from 192 sessions (21,089 tool calls) shows that Serena symbol lookups with file paths (`find_symbol`/`get_symbols_overview`) are frequently followed by an immediate `Read` of the same file path in-session: 102 of 554 calls (18.4%), with a common pattern of `find_symbol(include_body=true)` plus `Read(offset/limit)` context slicing.

mcp__serena__find_symbolmcp__serena__get_symbols_overviewClaude Code Readinclude_body

What Happened

A 21-day telemetry thread from 192 sessions (21,089 tool calls) shows that Serena symbol lookups with file paths (`find_symbol`/`get_symbols_overview`) are frequently followed by an immediate `Read` of the same file path in-session: 102 of 554 calls (18.4%), with a common pattern of `find_symbol(include_body=true)` plus `Read(offset/limit)` context slicing.
A 21-day telemetry thread from 192 sessions (21,089 tool calls) shows that Serena symbol lookups with file paths (`find_symbol`/`get_symbols_overview`) are frequently followed by an immediate `Read` of the same file path in-session: 102 of 554 calls (18.4%), with a common pattern of `find_symbol(include_body=true)` plus `Read(offset/limit)` context slicing.
1 evidence item attached for review.

What is Different

Before

Scattered source updates, isolated context, and manual follow-up across multiple feeds.

Now

The issue contributes concrete evidence that current symbol-query output is insufficient for downstream context needs in real usage, because agents still request additional surrounding lines even after `include_body=true`; the concrete direction is to make symbol responses include/optionally provide nearby context (e.g., `context_lines`) instead of forcing a second read step.

Why Track This

Why It Matters

Developers using Serena in Claude Code will feel symbol navigation as a two-step process rather than a single lookup, so they can experience slower edit loops and extra context-fetching overhead when nearby lines are missing. The signal is actionable because 18.4% of resolved symbol queries fallback to same-file reads, usually with `offset/limit`, so teams should validate whether richer symbol-context output lowers this fallback rate without increasing response size or noise.

Impact

What To Watch Next

Watch whether mcp__serena__find_symbol becomes a repeated pattern.
Track follow-up changes around AI Agents.
Compare future signals against this evidence trail.
Re-check risk flags: single_user_single_dataset_limits_generalization, fallback_reads_may_include_edit_tracking_workflow.

Open Topic Timeline Open Technical Event Open Original Sourcesingle_user_single_dataset_limits_generalization / fallback_reads_may_include_edit_tracking_workflow / context_enrichment_may_increase_payload_size

Supporting Evidence

GITHUB ISSUEHigh Trust

oraios/serena Issue #1491: Discussion: real-usage data on Serena adoption and `find_symbol` fallback patterns (n=21k tool calls, 192 sessions, 21 days)

Across the tracked data, 454 `find_symbol` calls had 52 follow-up reads (11.5%), and 169 `get_symbols_overview` calls had 50 follow-up reads (29.6%); median gap to fallback read was 30.9 seconds.