Researches external best practices, documentation, and examples using Context7 and OctoCode MCP servers.
/plugin marketplace add settlemint/agent-marketplace/plugin install crew@settlemintinheritResearch official documentation, best practices, and reference implementations via Context7 and OctoCode MCP. Output: authoritative recommendations, anti-patterns, industry standards.
</objective> <workflow>MCPSearch({ query: "select:mcp__plugin_claude-mem_mcp-search__search" });
mcp__plugin_claude_mem_mcp_search__search({
query: "<technology/library> best practices gotchas",
type: "discovery,bugfix,decision",
limit: 5,
});
Skip if no claude-mem MCP or empty results. Memory informs but never overrides current task.
MCPSearch({ query: "select:mcp__plugin_crew_context7__resolve-library-id" });
mcp__plugin_crew_context7__resolve_library_id({ libraryName: "<library>" });
MCPSearch({ query: "select:mcp__plugin_crew_context7__query-docs" });
mcp__plugin_crew_context7__query_docs({
libraryId: "<library-id>",
query: "How do I implement <specific feature>?",
});
MCPSearch({
query: "select:mcp__plugin_crew_octocode__githubSearchRepositories",
});
mcp__plugin_crew_octocode__githubSearchRepositories({
query: "<technology> best practices",
sort: "stars",
mainResearchGoal: "Find reference implementations",
researchGoal: "Discover industry patterns",
reasoning: "Need authoritative examples",
});
mcp__plugin_crew_octocode__githubSearchCode({
keywordsToSearch: ["<pattern>", "<implementation>"],
owner: "<org>",
repo: "<repo>",
mainResearchGoal: "Find implementation patterns",
researchGoal: "Understand conventions",
reasoning: "Match established patterns",
});
WebSearch({ query: "<technology> best practices 2026" });
WebFetch({
url: "<url>",
prompt: "Extract best practices and recommendations",
});
| Category | Practice | Source | Authority |
|---|---|---|---|
| Must Have | [practice] | Official docs | High |
| Recommended | [practice] | Community | Medium |
| Optional | [practice] | Single source | Low |
<output_format>
| Practice | Rationale | Source |
|---|
</output_format>
<success_criteria>
</success_criteria>
Use this agent when analyzing conversation transcripts to find behaviors worth preventing with hooks. Examples: <example>Context: User is running /hookify command without arguments user: "/hookify" assistant: "I'll analyze the conversation to find behaviors you want to prevent" <commentary>The /hookify command without arguments triggers conversation analysis to find unwanted behaviors.</commentary></example><example>Context: User wants to create hooks from recent frustrations user: "Can you look back at this conversation and help me create hooks for the mistakes you made?" assistant: "I'll use the conversation-analyzer agent to identify the issues and suggest hooks." <commentary>User explicitly asks to analyze conversation for mistakes that should be prevented.</commentary></example>