Cite as: Real Problem AI problem “Why can't you see which skill a subagent actually used when a task quietly goes wrong?”. Opportunity score 6.5 out of 10 (severity 6, AI feasibility 8, market signal 5, competition gap 7). Category AI / Agents. Trend Agents. Source signal: anthropics/claude-code GitHub issue #12142, PaulRBerg, 22 Nov 2025, closed as duplicate of #5974.. Canonical URL: https://www.realproblem.ai/archive/why-cant-you-see-which-skill-a-subagent-used-when-it-fails.
Why can't you see which skill a subagent actually used when a task quietly goes wrong?
When subagents, especially the Plan subagent, execute, there's no visibility into which skills they invoke, making it hard to diagnose which skill failed or discover what's available at all.
Who has it: Claude Code users building workflows around subagents and skills.
Evidence
“Lack of transparency: Users can't see what the subagent is actually doing under the hood... Difficult debugging: When things go wrong, it's hard to diagnose which skill failed or behaved unexpectedly.”
Quoted word for word from the public post linked below. Nobody submitted it to Real Problem AI.
anthropics/claude-code GitHub issue #12142, PaulRBerg, 22 Nov 2025, closed as duplicate of #5974.Why it is archived
Trimmed to 100-cap (lowest opportunity_score)
Scoring breakdown
Existing players
- Main-agent tool call display · Exists for the top-level agent but not propagated to subagent skill invocations.
- Verbose/debug logging · Buries the relevant skill-invocation signal in noise.
What they are missing
An inline display of skill invocations within subagent output, similar to tool-call visibility for the main agent, so failures can be traced to a specific skill instead of a black-box subagent run.
Stack hint
#ATB28 · Canonical URL: https://www.realproblem.ai/archive/why-cant-you-see-which-skill-a-subagent-used-when-it-fails