From project-toolkit
Gathers context from Forgetful Memory, Context7 docs, knowledge graph, and web sources for tasks or technologies, producing focused summaries with code snippets and insights.
How this command is triggered — by the user, by Claude, or both
Slash command
/project-toolkit:context_gather <task-or-technology>opusThis command is limited to the following tools:
The summary Claude sees in its command listing — used to decide when to auto-load this command
# Context Retrieval Command
**Purpose**: Gather relevant context from Forgetful Memory, Context7, and web sources before planning or implementing code.
**Usage**: `/context_gather <detailed task description>`
---
Use the **context-retrieval** subagent to gather context for the following task:
{TASK_DESCRIPTION}
The context-retrieval subagent will:
- Search Forgetful Memory across ALL projects for relevant patterns, decisions, and code
- Read linked code artifacts and documents
- Query Context7 for framework-specific guidance if applicable
- Explore the knowledge graph to find connect...Purpose: Gather relevant context from Forgetful Memory, Context7, and web sources before planning or implementing code.
Usage: /context_gather <detailed task description>
Use the context-retrieval subagent to gather context for the following task:
{TASK_DESCRIPTION}
The context-retrieval subagent will:
Wait for the subagent to return its findings before proceeding.
npx claudepluginhub rjmurillo/ai-agents --plugin project-toolkit/context_gatherGather comprehensive context from Forgetful Memory, Context7 docs, and web sources before planning or implementation
/analyze-contextRetrieves project context and relevant memories, then synthesizes goals, decisions, gotchas, and prior insights into a pre-task brief.
/contextSearches project and global memories by keyword with smart expansion to related terms (e.g., auth→JWT/OAuth, db→Prisma/PostgreSQL), displaying matching key-value entries.
/orchestrateActivates memory orchestrator mode for complex tasks: decomposes into subtasks, delegates to subagents, updates persistent memory files with tags and context.
/recallSearches the KongCode memory graph for past knowledge matching a query and displays results with source types, dates, and relevance scores.
/primeLoads relevant knowledge base facts (rules, gotchas, patterns, decisions) into context before starting work. Also supports auto-priming on session start, file touch, and keyword detection.