Code graph and memory for your coding agents, in one Rust program. For Claude Code, Codex, Cursor and any MCP client.
Rust agent-memoryai-agentsclaude-codecode-analysiscode-graphcode-intelligencecode-searchcodexcoding-agentscontext-engineeringcursordeveloper-toolsgraphragknowledge-graphlocal-firstmcpmcp-servermodel-context-protocolrusttree-sitter
{
"_SEARCH_NUDGE": "{\"hookSpecificOutput\":{\"hookEventName\":\"PreToolUse\",\"additionalContext\":\"MANDATORY: .graphicle/graph.json exists. You MUST run `graphicle-graph query \\\"<question>\\\"` before grepping raw files. Grep raw files only once you know the exact file and lines you need, from the graph or an earlier read.\"}}\n",
"_READ_NUDGE": "{\"hookSpecificOutput\":{\"hookEventName\":\"PreToolUse\",\"additionalContext\":\"MANDATORY: .graphicle/graph.json exists. You MUST run graphicle-graph before reading source files. Use: `graphicle-graph query \\\"<question>\\\"` (scoped subgraph), `graphicle-graph explain \\\"<concept>\\\"`, or `graphicle-graph path \\\"<A>\\\" \\\"<B>\\\"`. Read raw files only once graphicle-graph has shown you the exact file and lines you need. This rule applies to subagents too — include it in every subagent prompt involving code exploration.\"}}\n",
"_READ_NUDGE_STALE": "{\"hookSpecificOutput\":{\"hookEventName\":\"PreToolUse\",\"additionalContext\":\".graphicle/graph.json exists but may be STALE for this file (the file changed after the last build). Prefer `graphicle-graph query \\\"<question>\\\"` for orientation, and run `graphicle-graph update` to refresh the graph. Reading the file directly is fine.\"}}\n",
"_READ_DENY": "{\"hookSpecificOutput\":{\"hookEventName\":\"PreToolUse\",\"permissionDecision\":\"deny\",\"permissionDecisionReason\":\"graphicle strict mode: this project has a fresh knowledge graph that covers this file. Run `graphicle-graph query \\\"<your question>\\\"` (or `graphicle-graph explain` / `graphicle-graph path`) FIRST to orient yourself, then re-issue this Read — it will be allowed. This block fires at most once per session; after one query, read raw files for the exact lines the graph pointed you to. Apply the same rule in any subagent prompt that explores code.\"}}\n",
"_HOOK_SOURCE_EXTS": [
".py",
".js",
".cjs",
".ts",
".tsx",
".jsx",
".astro",
".vue",
".svelte",
".ejs",
".go",
".rs",
".java",
".rb",
".c",
".h",
".cpp",
".hpp",
".cc",
".cs",
".kt",
".swift",
".php",
".scala",
".lua",
".ps1",
".psm1",
".psd1",
".sh",
".md",
".rst",
".txt",
".mdx"
],
"_GEMINI_NUDGE_TEXT": "graphicle: knowledge graph at .graphicle/. For focused questions, run `graphicle-graph query \"<question>\"` (scoped subgraph, usually much smaller than GRAPH_REPORT.md) instead of grepping raw files. Read GRAPH_REPORT.md only for broad architecture context."
}