Step 225: Claude tool definitions

PROGRESS.md

@@ -740,6 +740,7 @@ vcpkg's imgui 1.91.9 removed the `sdl2-binding` feature (only `sdl3-binding` exi
 | 2026-02-10 | Codex | Step 222: Evaluation workflow task suite (20 multi-step tasks). 1/1 tests pass (step222_test). |
 | 2026-02-10 | Codex | Step 223: Evaluation runner CLI (whetstone_eval, task/trace/report pipeline). 1/1 tests pass (step223_test). |
 | 2026-02-10 | Codex | Step 224: Evaluation harness tests (good/bad/partial scoring). 1/1 tests pass (step224_test). |
+| 2026-02-10 | Codex | Step 225: Claude tool definitions (Anthropic tool schema + system prompt). 1/1 tests pass (step225_test). |
 | 2026-02-10 | Codex | Step 166: Extract main.cpp into panel headers marked complete (already implemented). step166_test passes; file_limits_test 4/4 passes. |
 | 2026-02-10 | Claude Opus 4.6 | Sprint 7 Phase 7a (Steps 202–206): API docs (AGENT_API.md, 23 methods), JSON schemas (20 files), exposed ContextAPI/BatchMutationAPI/Pipeline via RPC, permission policy updates. 51/51 tests pass. |
 | 2026-02-10 | Claude Opus 4.6 | Sprint 7 Phase 7b (Steps 207–213): MCPServer.h (10 tools, 5 resources, 4 prompts, JSON-RPC 2.0 initialize handshake), MCPBridge.h (stdio transport). 90/90 tests pass. |

editor/CMakeLists.txt

@@ -1158,6 +1158,9 @@ target_include_directories(step223_test PRIVATE src)
 add_executable(step224_test tests/step224_test.cpp)
 target_include_directories(step224_test PRIVATE src)
 target_link_libraries(step224_test PRIVATE nlohmann_json::nlohmann_json)
+
+add_executable(step225_test tests/step225_test.cpp)
+target_include_directories(step225_test PRIVATE src)
 add_executable(step213_test tests/step213_test.cpp)
 target_include_directories(step213_test PRIVATE src)
 target_link_libraries(step213_test PRIVATE nlohmann_json::nlohmann_json)

editor/tests/step225_test.cpp

@@ -0,0 +1,34 @@
+// Step 225: Claude tool definitions.
+
+#include <cassert>
+#include <fstream>
+#include <string>
+
+static std::string readFile(const std::string& path) {
+    std::ifstream f(path);
+    if (!f.is_open()) return {};
+    return std::string((std::istreambuf_iterator<char>(f)),
+                       std::istreambuf_iterator<char>());
+}
+
+static void assertContains(const std::string& text, const std::string& needle) {
+    assert(text.find(needle) != std::string::npos);
+}
+
+int main() {
+    const std::string tools = readFile("../tools/claude/tools.json");
+    const std::string prompt = readFile("../tools/claude/system_prompt.txt");
+    assertContains(tools, "whetstone_get_ast");
+    assertContains(tools, "whetstone_mutate");
+    assertContains(tools, "whetstone_batch_mutate");
+    assertContains(tools, "whetstone_get_scope");
+    assertContains(tools, "whetstone_get_call_hierarchy");
+    assertContains(tools, "whetstone_suggest_annotations");
+    assertContains(tools, "whetstone_apply_annotation");
+    assertContains(tools, "whetstone_generate_code");
+    assertContains(tools, "whetstone_run_pipeline");
+    assertContains(tools, "whetstone_project_language");
+    assertContains(prompt, "Whetstone coding assistant");
+    printf("step225_test: all assertions passed\n");
+    return 0;
+}

sprint7_plan.md

@@ -347,7 +347,7 @@ Test suites that measure how accurately an LLM can use Whetstone tools.
 
 Optimized tool definitions and prompts for specific LLM families.
 
-- [ ] **Step 225: Claude tool definitions**
+- [x] **Step 225: Claude tool definitions**
   Anthropic-format tool definitions optimized for Claude:
   - Each Whetstone MCP tool as a Claude `tool_use` definition
   - Descriptions tuned for Claude's strengths (detailed, structured)

tools/claude/system_prompt.txt

@@ -0,0 +1,12 @@
+You are a Whetstone coding assistant with access to semantic annotations and AST tools.
+
+Guidelines:
+- Use tools to read or mutate the AST instead of guessing structure.
+- Prefer safe, minimal mutations. Validate after multi-step changes.
+- If asked to annotate memory strategy, call whetstone_suggest_annotations then
+  whetstone_apply_annotation with the highest-confidence suggestion.
+- When generating code from a spec, call whetstone_generate_code and explain any assumptions.
+- For refactors, use whetstone_get_ast to locate targets, then whetstone_mutate or
+  whetstone_batch_mutate. Re-run whetstone_run_pipeline if asked to validate.
+
+You must follow user instructions carefully. If a step is ambiguous, ask a brief clarifying question.

tools/claude/tools.json

@@ -0,0 +1,164 @@
+{
+  "version": "1.0",
+  "provider": "anthropic",
+  "tools": [
+    {
+      "name": "whetstone_get_ast",
+      "description": "Get the current AST (Abstract Syntax Tree) of the active buffer as JSON.",
+      "input_schema": {
+        "type": "object",
+        "properties": {},
+        "additionalProperties": false
+      },
+      "examples": [
+        {}
+      ]
+    },
+    {
+      "name": "whetstone_mutate",
+      "description": "Apply a single mutation to the AST (setProperty, updateNode, deleteNode, insertNode).",
+      "input_schema": {
+        "type": "object",
+        "properties": {
+          "type": {
+            "type": "string",
+            "enum": ["setProperty", "updateNode", "deleteNode", "insertNode"]
+          },
+          "nodeId": { "type": "string" },
+          "property": { "type": "string" },
+          "value": { "type": "string" },
+          "parentId": { "type": "string" },
+          "role": { "type": "string" },
+          "node": { "type": "object" }
+        },
+        "required": ["type"]
+      },
+      "examples": [
+        { "type": "setProperty", "nodeId": "fn1", "property": "name", "value": "rename_me" }
+      ]
+    },
+    {
+      "name": "whetstone_batch_mutate",
+      "description": "Apply multiple mutations atomically (all succeed or rollback).",
+      "input_schema": {
+        "type": "object",
+        "properties": {
+          "mutations": {
+            "type": "array",
+            "items": { "type": "object" }
+          }
+        },
+        "required": ["mutations"]
+      },
+      "examples": [
+        { "mutations": [{ "type": "deleteNode", "nodeId": "n42" }] }
+      ]
+    },
+    {
+      "name": "whetstone_get_scope",
+      "description": "Get all symbols in scope at a given AST node.",
+      "input_schema": {
+        "type": "object",
+        "properties": {
+          "nodeId": { "type": "string" }
+        },
+        "required": ["nodeId"]
+      },
+      "examples": [
+        { "nodeId": "fn1" }
+      ]
+    },
+    {
+      "name": "whetstone_get_call_hierarchy",
+      "description": "Get call hierarchy for a function node (callers and callees).",
+      "input_schema": {
+        "type": "object",
+        "properties": {
+          "functionId": { "type": "string" }
+        },
+        "required": ["functionId"]
+      },
+      "examples": [
+        { "functionId": "fn1" }
+      ]
+    },
+    {
+      "name": "whetstone_suggest_annotations",
+      "description": "Get memory annotation suggestions for a node or line/col location.",
+      "input_schema": {
+        "type": "object",
+        "properties": {
+          "nodeId": { "type": "string" },
+          "line": { "type": "integer" },
+          "col": { "type": "integer" }
+        }
+      },
+      "examples": [
+        { "nodeId": "fn1" }
+      ]
+    },
+    {
+      "name": "whetstone_apply_annotation",
+      "description": "Apply a memory annotation suggestion to the AST.",
+      "input_schema": {
+        "type": "object",
+        "properties": {
+          "nodeId": { "type": "string" },
+          "annotationType": { "type": "string" },
+          "strategy": { "type": "string" },
+          "reason": { "type": "string" },
+          "confidence": { "type": "number" }
+        },
+        "required": ["nodeId", "annotationType", "strategy"]
+      },
+      "examples": [
+        { "nodeId": "fn1", "annotationType": "ReclaimAnnotation", "strategy": "Tracing" }
+      ]
+    },
+    {
+      "name": "whetstone_generate_code",
+      "description": "Generate code from a natural language spec, optionally preferring imports.",
+      "input_schema": {
+        "type": "object",
+        "properties": {
+          "spec": { "type": "string" },
+          "preferImports": { "type": "boolean" }
+        },
+        "required": ["spec"]
+      },
+      "examples": [
+        { "spec": "Create a function that sums a list of integers.", "preferImports": true }
+      ]
+    },
+    {
+      "name": "whetstone_run_pipeline",
+      "description": "Run full pipeline: parse, infer annotations, validate, optimize, generate.",
+      "input_schema": {
+        "type": "object",
+        "properties": {
+          "source": { "type": "string" },
+          "sourceLanguage": { "type": "string" },
+          "targetLanguage": { "type": "string" }
+        },
+        "required": ["source", "sourceLanguage", "targetLanguage"]
+      },
+      "examples": [
+        { "source": "def add(a,b): return a+b", "sourceLanguage": "python", "targetLanguage": "cpp" }
+      ]
+    },
+    {
+      "name": "whetstone_project_language",
+      "description": "Project current AST to a different target language.",
+      "input_schema": {
+        "type": "object",
+        "properties": {
+          "targetLanguage": { "type": "string" }
+        },
+        "required": ["targetLanguage"]
+      },
+      "examples": [
+        { "targetLanguage": "rust" }
+      ]
+    }
+  ]
+}