feat(graphify): add graphify knowledge-graph plugin

.agents/plugins/marketplace.json

@@ -699,6 +699,18 @@
         "authentication": "ON_INSTALL"
       },
       "category": "Productivity"
+    },
+    {
+      "name": "graphify",
+      "source": {
+        "source": "local",
+        "path": "./plugins/graphify"
+      },
+      "policy": {
+        "installation": "AVAILABLE",
+        "authentication": "ON_INSTALL"
+      },
+      "category": "Tooling"
     }
   ]
 }

.claude-plugin/marketplace.json

@@ -675,6 +675,14 @@
       "keywords": ["skillopt", "sleep", "self-improvement", "memory-consolidation", "skills", "continual-learning"],
       "tags": ["productivity", "skills", "hooks"],
       "source": "./plugins/skillopt-sleep"
+    },
+    {
+      "name": "graphify",
+      "description": "Turn any folder of code, docs, papers, images, or videos into a queryable knowledge graph with community detection, query/path/explain tools, and an honest audit trail",
+      "category": "tooling",
+      "keywords": ["graphify", "knowledge-graph", "graphrag", "codebase", "code-understanding", "ast", "community-detection"],
+      "tags": ["skills", "knowledge-graph"],
+      "source": "./plugins/graphify"
     }
   ]
 }

plugins/ast-grep/hooks.json

@@ -7,7 +7,7 @@
         "hooks": [
           {
             "type": "command",
-            "command": "${CLAUDE_PLUGIN_ROOT}/hooks/context.sh",
+            "command": "./hooks/context.sh",
             "timeout": 10
           }
         ]

plugins/fetch/hooks.json

@@ -6,7 +6,7 @@
         "hooks": [
           {
             "type": "command",
-            "command": "${CLAUDE_PLUGIN_ROOT}/hooks/webfetch-fallback.sh",
+            "command": "./hooks/webfetch-fallback.sh",
             "timeout": 5
           }
         ]

plugins/gatekeeper/hooks.json

@@ -7,7 +7,7 @@
         "hooks": [
           {
             "type": "command",
-            "command": "node ${CLAUDE_PLUGIN_ROOT}/dist/pre-tool-use.js",
+            "command": "node ./dist/pre-tool-use.js",
             "timeout": 5
           }
         ]

plugins/google-workspace/hooks.json

@@ -6,7 +6,7 @@
         "hooks": [
           {
             "type": "command",
-            "command": "${CLAUDE_PLUGIN_ROOT}/hooks/context.sh",
+            "command": "./hooks/context.sh",
             "timeout": 10
           }
         ]

plugins/graphify/.claude-plugin/plugin.json

@@ -0,0 +1,23 @@
+{
+  "name": "graphify",
+  "version": "1.0.0",
+  "description": "Turn any folder of code, docs, papers, images, or videos into a queryable knowledge graph with community detection, query/path/explain tools, and an honest audit trail",
+  "author": {
+    "name": "Safi Shamsi",
+    "url": "https://github.com/safishamsi"
+  },
+  "homepage": "https://github.com/safishamsi/graphify",
+  "repository": "https://github.com/safishamsi/graphify",
+  "license": "MIT",
+  "keywords": [
+    "graphify",
+    "knowledge-graph",
+    "graphrag",
+    "codebase",
+    "code-understanding",
+    "ast",
+    "community-detection",
+    "obsidian",
+    "neo4j"
+  ]
+}

plugins/graphify/.codex-plugin/plugin.json

@@ -0,0 +1,37 @@
+{
+  "name": "graphify",
+  "version": "1.0.0",
+  "description": "Turn any folder of code, docs, papers, images, or videos into a queryable knowledge graph with community detection, query/path/explain tools, and an honest audit trail",
+  "author": {
+    "name": "Safi Shamsi",
+    "url": "https://github.com/safishamsi"
+  },
+  "interface": {
+    "displayName": "Graphify",
+    "shortDescription": "Turn any folder of code, docs, papers, images, or videos into a queryable knowledge graph with community detection, query/path/explain tools, and an honest a...",
+    "longDescription": "Turn any folder of code, docs, papers, images, or videos into a queryable knowledge graph with community detection, query/path/explain tools, and an honest audit trail",
+    "developerName": "Safi Shamsi",
+    "category": "Tooling",
+    "capabilities": [
+      "Skill"
+    ],
+    "defaultPrompt": [
+      "Help me use Graphify for my current task."
+    ],
+    "websiteURL": "https://github.com/safishamsi/graphify"
+  },
+  "homepage": "https://github.com/safishamsi/graphify",
+  "repository": "https://github.com/safishamsi/graphify",
+  "license": "MIT",
+  "keywords": [
+    "graphify",
+    "knowledge-graph",
+    "graphrag",
+    "codebase",
+    "code-understanding",
+    "ast",
+    "community-detection",
+    "obsidian",
+    "neo4j"
+  ]
+}

plugins/graphify/README.md

@@ -0,0 +1,47 @@
+# Graphify
+
+Turn any folder of code, docs, papers, images, or videos into a navigable **knowledge graph** with community detection, an honest audit trail (EXTRACTED / INFERRED / AMBIGUOUS), and three outputs: interactive HTML, GraphRAG-ready JSON, and a plain-language `GRAPH_REPORT.md`.
+
+This plugin bundles the official Graphify Claude skill, so Claude automatically treats questions about a codebase — its architecture, file relationships, or project content — as graph queries, and can build the graph on demand.
+
+## Installation
+
+```sh
+claude
+/plugin marketplace add pleaseai/claude-code-plugins
+/plugin install graphify@pleaseai
+```
+
+The skill installs the underlying `graphifyy` CLI on first use (via `uv tool install` or `pip`), so no manual setup is required. To install it yourself ahead of time:
+
+```sh
+uv tool install graphifyy   # recommended
+# or: pipx install graphifyy
+```
+
+## Usage
+
+Once installed, ask Claude natural-language questions about a project, or invoke the pipeline directly:
+
+```
+/graphify                      # build a graph for the current directory
+/graphify ./docs --update      # re-extract only changed files
+/graphify query "describe the auth flow"
+/graphify path "UserService" "DatabasePool"
+/graphify explain "SwinTransformer"
+```
+
+When `graphify-out/graph.json` already exists, plain questions like "How does X work?" are answered straight from the graph.
+
+## Optional environment variables
+
+| Variable | Purpose |
+|----------|---------|
+| `GEMINI_API_KEY` / `GOOGLE_API_KEY` | Use Gemini for semantic extraction instead of in-session subagents |
+| `GRAPHIFY_GEMINI_MODEL` | Override the default Gemini model |
+| `GRAPHIFY_MAX_WORKERS` | AST parallelism thread count |
+
+## Links
+
+- Source: https://github.com/safishamsi/graphify
+- License: MIT

plugins/graphify/hooks.json

@@ -0,0 +1,17 @@
+{
+  "description": "Load graphify usage rules at session start when the project has a knowledge graph (graphify-out/graph.json)",
+  "hooks": {
+    "SessionStart": [
+      {
+        "matcher": "startup",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "./hooks/context.sh",
+            "timeout": 10
+          }
+        ]
+      }
+    ]
+  }
+}

plugins/graphify/hooks/claude-md.md

@@ -0,0 +1,9 @@
+## graphify
+
+This project has a knowledge graph at graphify-out/ with god nodes, community structure, and cross-file relationships.
+
+Rules:
+- For codebase questions, first run `graphify query "<question>"` when graphify-out/graph.json exists. Use `graphify path "<A>" "<B>"` for relationships and `graphify explain "<concept>"` for focused concepts. These return a scoped subgraph, usually much smaller than GRAPH_REPORT.md or raw grep output.
+- If graphify-out/wiki/index.md exists, use it for broad navigation instead of raw source browsing.
+- Read graphify-out/GRAPH_REPORT.md only for broad architecture review or when query/path/explain do not surface enough context.
+- After modifying code, run `graphify update .` to keep the graph current (AST-only, no API cost).

plugins/graphify/hooks/context.sh

@@ -0,0 +1,29 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# Only inject graphify's always-on rules when this project actually has a graph.
+# The context describes how to use graphify-out/, so it is irrelevant (and just
+# noise) in sessions for projects that have never been graphified.
+if [ ! -f "graphify-out/graph.json" ]; then
+  exit 0
+fi
+
+# Resolve the plugin root from CLAUDE_PLUGIN_ROOT, falling back to this script's
+# own location when the runtime does not set it (e.g. Antigravity). Without the
+# fallback, `set -u` would abort here with "unbound variable".
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+CONTEXT_FILE="${CLAUDE_PLUGIN_ROOT:-$SCRIPT_DIR/..}/hooks/claude-md.md"
+[ -f "$CONTEXT_FILE" ] || exit 0
+CONTEXT_CONTENT=$(cat "$CONTEXT_FILE")
+
+if command -v jq >/dev/null 2>&1; then
+  jq -n --arg context "$CONTEXT_CONTENT" '{
+    "hookSpecificOutput": {
+      "hookEventName": "SessionStart",
+      "additionalContext": $context
+    }
+  }'
+else
+  # Fallback: SessionStart hooks also accept raw stdout as additional context.
+  printf '%s\n' "$CONTEXT_CONTENT"
+fi

plugins/graphify/hooks/hooks.json

@@ -0,0 +1,17 @@
+{
+  "description": "Load graphify usage rules at session start when the project has a knowledge graph (graphify-out/graph.json)",
+  "hooks": {
+    "SessionStart": [
+      {
+        "matcher": "startup",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "${CLAUDE_PLUGIN_ROOT}/hooks/context.sh",
+            "timeout": 10
+          }
+        ]
+      }
+    ]
+  }
+}

plugins/graphify/plugin.json

@@ -0,0 +1,23 @@
+{
+  "name": "graphify",
+  "version": "1.0.0",
+  "description": "Turn any folder of code, docs, papers, images, or videos into a queryable knowledge graph with community detection, query/path/explain tools, and an honest audit trail",
+  "author": {
+    "name": "Safi Shamsi",
+    "url": "https://github.com/safishamsi"
+  },
+  "homepage": "https://github.com/safishamsi/graphify",
+  "repository": "https://github.com/safishamsi/graphify",
+  "license": "MIT",
+  "keywords": [
+    "graphify",
+    "knowledge-graph",
+    "graphrag",
+    "codebase",
+    "code-understanding",
+    "ast",
+    "community-detection",
+    "obsidian",
+    "neo4j"
+  ]
+}