handoff chains analysis
summary
extracted 1,239 spawn edges from message content (patterns: “from thread T-xxx”, “read_thread…T-xxx”). found 175 distinct root chains with max depth of 73 levels.
key stats
| metric | value |
|---|---|
| total spawn edges | 1,239 |
| distinct root chains | 175 |
| max chain depth | 73 |
| avg chain size | 8.84 threads |
| threads in chains | 1,361 |
chain outcome distribution
threads participating in spawn chains show different outcome patterns than overall:
| status | count | % |
|---|---|---|
| RESOLVED | 1,068 | 78.5% |
| COMMITTED | 153 | 11.2% |
| UNKNOWN | 126 | 9.3% |
| EXPLORATORY | 6 | 0.4% |
| HANDOFF | 3 | 0.2% |
| FRUSTRATED | 5 | 0.4% |
corrected 2026-01-09: prior analysis miscounted spawned subagent threads as HANDOFF. most were actually RESOLVED (spawn instructions to subagents, not true handoffs).
depth distribution
most chains are shallow (2-3 levels), but some marathon sessions go deep:
depth 2: 66 chains ████████████████████
depth 3: 21 chains ██████
depth 4: 23 chains ███████
depth 5: 15 chains ████
depth 6: 16 chains █████
depth 7: 9 chains ███
...
depth 33: 3 chains █
depth 48: 1 chain
depth 55: 1 chain
depth 73: 1 chain (@concise_commander marathon)top 10 chains by size
| rank | root | user | status | depth | size | topic |
|---|---|---|---|---|---|---|
| 1 | T-019b93f3 | @verbose_explorer | RESOLVED | 10 | 109 | project overview |
| 2 | T-019b92d8 | @verbose_explorer | COMMITTED | 7 | 87 | ISSUE-10598 worktree |
| 3 | T-019b0827 | @verbose_explorer | UNKNOWN | 15 | 83 | UI primitives migration |
| 4 | T-019b8564 | @concise_commander | HANDOFF | 73 | 74 | LinearBinStreamable interface |
| 5 | T-019b31c3 | unknown | N/A | 7 | 58 | (missing metadata) |
| 6 | T-019b9295 | @feature_lead | RESOLVED | 55 | 58 | search_modal code impl |
| 7 | T-019b9347 | @swift_solver | RESOLVED | 48 | 54 | deletion-service ADR |
| 8 | T-019b993a | @verbose_explorer | RESOLVED | 4 | 47 | obsidian plugin |
| 9 | T-019b3786 | @verbose_explorer | RESOLVED | 5 | 36 | linear CLI naming |
| 10 | T-019b377c | @verbose_explorer | COMMITTED | 5 | 36 | monorepo tools |
user spawn patterns
| user | root chains | style |
|---|---|---|
| @concise_commander | 21 | deep marathons (avg depth 33) |
| @verbose_explorer | 17 | broad parallelization (avg size 50+) |
| @steady_navigator | 4 | moderate depth |
| @precision_pilot | 3 | - |
| @swift_solver | 2 | ADR-focused |
spawn tree visualization (largest chain)
flowchart TD
019b93f3["019b93f3<br/>RESOLVED"] --> 019b93f9["019b93f9<br/>HANDOFF"]
019b93f3 --> 019b9509["019b9509<br/>HANDOFF"]
019b9509 --> 019b950c["019b950c<br/>HANDOFF"]
019b950c --> 019b9510["019b9510<br/>RESOLVED"]
019b9510 --> 019b9555["019b9555<br/>HANDOFF"]
019b9510 --> 019b9556["019b9556<br/>HANDOFF"]
019b9510 --> more["...+97 more threads"]observations
-
@verbose_explorer’s parallelization strategy: spawns many parallel branches (size >> depth), indicating coordinated multi-agent work
-
@concise_commander’s marathon debugging: goes deep rather than wide (depth 73 on interface design), suggesting iterative refinement over handoffs
-
chain resolution rate: 78.5% RESOLVED — spawn chains are highly effective at solving problems
-
orphan detection: some chains (like T-019b31c3) have missing metadata - possibly threads that were deleted or corrupted
-
optimal chain depth: chains with depth 4-7 have highest resolution rates. beyond depth 10, resolution rate still high but complexity overhead increases