| name | team-shinchan:analyze |
| description | Use when you need deep analysis of code, bugs, performance, or architecture issues. |
| user-invocable | false |
EXECUTE IMMEDIATELY
Step 1: Validate Input
If args is empty or only whitespace:
Ask user: "What would you like to analyze?"
STOP and wait for user response
If args length > 2000 characters:
Truncate to 2000 characters
Warn user: "Request was truncated to 2000 characters"
Step 1c: Single-pass opt-out
If args contains the flag --single, skip the panel and run the Fallback single-Hiroshi Task at the bottom. Otherwise run the diverse-lens panel. (Match the literal flag only — do NOT substring-match words like "quick", which legitimately appear in targets such as "analyze the quicksort path".)
Step 2: Diverse-lens analysis panel (default)
Quality-first: a single analyzer sees one frame and systematically misses what that frame can't see (a structural pass overlooks the auth edge case; a correctness pass overlooks the architectural smell). Launch independent read-only lenses in PARALLEL, then synthesize.
Task(subagent_type="team-shinchan:hiroshi", model="opus",
prompt=`/team-shinchan:analyze (architecture & root-cause lens). READ-ONLY: do not modify any file.
Analyze: ${args}
Focus: structure, dependencies, design seams; if a bug, the most likely root cause with evidence.
Return concrete, file:line-anchored findings + recommended direction.`)
(subagent_type=, model=,
prompt=)
(subagent_type=, model=,
prompt=)
