From unstuck
Use when entering Strike 3 of the unstuck protocol — two targeted attempts have failed and the issue needs broader rethink or user escalation
npx claudepluginhub jugrajsingh/skillgarden --plugin unstuckThis skill is limited to using the following tools:
Strike 3 of the unstuck protocol. Two targeted attempts failed — broaden the search and involve the user.
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).
Generates original PNG/PDF visual art via design philosophy manifestos for posters, graphics, and static designs on user request.
Strike 3 of the unstuck protocol. Two targeted attempts failed — broaden the search and involve the user.
Two strikes failed. Something foundational may be wrong:
Try WebSearch for the exact error message and relevant context:
WebSearch: "{EXACT_ERROR_MESSAGE} {FRAMEWORK} {LANGUAGE}"
If WebSearch is unavailable (deferred tool not loaded), ask the user to search manually:
AskUserQuestion:
question: "I can't search the web directly. Could you search for this and paste relevant results?"
header: "Web Search"
options:
- label: "I'll search now"
description: "Search: {EXACT_ERROR_MESSAGE} {FRAMEWORK}"
- label: "Skip external search"
description: "Continue with what we have"
Check (from results or user-provided context):
If a knowledge pack was loaded:
Compile everything learned across all three strikes:
What we know:
- Strike 1 ruled out: {LIST}
- Strike 2 ruled out: {LIST}
- External search found: {FINDINGS}
- Remaining hypotheses: {LIST}
Present a structured escalation:
AskUserQuestion:
question: "I've hit a wall after 3 investigation attempts. Here's what I found — can you provide guidance?"
header: "Escalation"
options:
- label: "Share context"
description: "I have additional context that might help (env details, recent changes, known issues)"
- label: "Try hypothesis"
description: "Pick from the remaining hypotheses and test it"
- label: "Different direction"
description: "Abandon this approach entirely, suggest an alternative"
- label: "Accept workaround"
description: "Use a temporary workaround while the root cause is investigated later"
Before asking, output the full context:
## Escalation Report
### Problem
{PROBLEM_SUMMARY}
### What Was Tried
Strike 1 (Diagnose): {APPROACH} → {OUTCOME}
Strike 2 (Pivot): {APPROACH} → {OUTCOME}
### Errors Encountered
{SPECIFIC_ERRORS}
### Remaining Hypotheses
{NUMBERED_LIST}
### External References
{URLS_OR_FINDINGS}
Based on user response:
Report to the orchestrator:
Escalation:
User guidance: {SUMMARY}
Action taken: {WHAT_WAS_DONE}
Result: { resolved | workaround | deferred }
Notes: {ADDITIONAL_CONTEXT}