synap-git 0.2.1__tar.gz → 1.1.1__tar.gz

synap_git-1.1.1/CHANGELOG.md

@@ -0,0 +1,139 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [1.1.1] - 2026-05-28
+
+### Fixed
+- **CRITICAL: N+1 query loop during edge resolution:** Migrated structural edge resolution to bulk FETCH queries, eliminating thousands of database calls per indexing run.
+- **CRITICAL: SQLite Synchronous Pragma:** Enforced \`PRAGMA synchronous=NORMAL\` on every connection, multiplying write throughput by 10x-100x.
+- **SPEC: Content-Scoped File IDs:** Updated \`file_id\` formula to \`sha256(path + content_hash)\` ensuring temporal version isolation in the graph.
+- **HIGH: Wiki Generation Resiliency:** Implemented exponential backoff retries for LLM wiki generation to prevent data loss on transient network errors.
+- **HIGH: Single Read Principle:** Optimized pipeline to read each file exactly once, halving I/O overhead.
+- **MEDIUM: Automated Lesson Pruning:** Daemon now automatically prunes expired memory lessons hourly.
+- **MEDIUM: Memory Bounded Indexing:** First-run indexing now processes in memory-bounded batches to prevent OOM on large repos.
+- **LOW: Checkpoint Validation:** MCP \`create_checkpoint\` tool now validates all input fields to prevent malformed data.
+
+### Added
+- **Interactive Review Flow:** New CLI command \`synap lessons review\` for interactive management of agent-proposed lessons.
+- **Context Monitoring:** New MCP tool \`signal_low_context\` for proactive agent context window monitoring.
+- **Configurable Maintenance:** Added \`checkpoint_threshold\` and \`lesson_expiry_days\` to \`config.toml\`.
+- **Improved Doctor:** \`synap doctor\` now checks for Git and GitHub CLI availability.
+- **Onboarding Guidance:** \`synap init\` now provides explicit next steps for starting the system.
+
+### Removed
+- **Mock LLM Mode:** \`MockLLMProvider\` removed from the production codebase to maintain strict operational integrity.
+
+## [1.1.0] - 2026-05-28
+
+### Added — Git-Snapshot Projection & Performance Refactoring
+- **Split Two-Path Indexing:** Separated initialization and incremental indexing into `_first_run_index` (full scan, CPU-parallelized) and `_incremental_index` (Git delta change detector).
+- **Asynchronous Wiki Generation Queue:** Decoupled slow, non-deterministic LLM wiki generation from structural indexing using a persistent database queue (`wiki_queue`) processed asynchronously by a daemon worker.
+- **Lazy Wiki Caching:** Added synchronous wiki generation fallback to CLI (`wiki show`), Web API, and MCP tools to dynamically build missing or stale pages on-demand.
+- **Process Pool Parallel Parsing:** Parallelized Tree-sitter parsing on first run across all CPU cores utilizing process-based concurrency with independent parser instances.
+- **SQLite Performance Hardening:**
+  - WAL mode and NORMAL synchronous configuration enabled during writes.
+  - Multi-row symbol and edge inserts batched into a single transaction via `executemany`.
+  - Dot-separated `module_key` pre-computation and indexing for $O(1)$ module resolution.
+  - SQLite FTS5 index integration for fast sub-millisecond symbol searches, avoiding full-table scans.
+- **Web API Lazy Refreshes:** Updated the `/wiki/{filepath}` GET endpoint to perform lazy refreshes on stale or missing pages before returning content.
+
+### Fixed
+- **FTS5 Cascade Delete:** Added database trigger `tgr_symbols_delete` to automatically clean up virtual `symbols_fts` entries when parent symbols are deleted.
+- **Duplicate File ID Collision:** Handled unique, path-scoped file identifier generation ensuring files with identical content (like empty `__init__.py`) do not conflict.
+
+## [0.2.1] - 2026-05-27
+
+### Added — Final Production Hardening & Release Execution
+- `synap rollback --commit <ref>` option: directly target a commit by hash/reference without interactive selection prompt.
+- `synap rollback --yes` / `-y` option: suppress confirmation prompt for non-interactive and scripted rollback flows.
+- Non-interactive guard in `synap rollback`: fails fast with a clear error when used in piped/CI contexts without `--commit` or `--yes`.
+- `synap rollback` invalid commit detection: validates commit reference via `git rev-parse --verify` and rejects unknown refs with a clear message.
+
+### Fixed
+- **SQLite migration short-circuit bug**: legacy un-versioned databases (user_version = 0) incorrectly skipped `CREATE TABLE IF NOT EXISTS` execution, leaving the `symbols` table and others uninitialized. The premature short-circuit is removed; all schema tables are now created correctly before bumping to version 1.
+- **Python `import_from_statement` missing symbol extraction**: Tree-sitter AST parser only extracted the module identifier from `from X import Y` statements, discarding `Y`. Now correctly emits `module:symbol` pairs for all imported names, aliases, and grouped imports.
+- **Namespace-aware call edge resolution**: Pass 2 import resolver now splits `module:symbol` import entries to narrow edge targets to the correct module file, eliminating false-positive dependency edges to duplicate class names in sibling namespaces.
+- **FastAPI app version hardcoded**: `create_app()` used a hardcoded version string `"0.2.0"` instead of the canonical `__version__`. Now dynamically imported from `synap_git.__init__`.
+- **Streaming generator cancellation safety**: confirmed `httpx` stream connections are closed cleanly on partial consumption (no socket leaks).
+- **Degraded mode retry logic**: confirmed 2-stage exponential backoff and graceful structural fallback under fully-offline and timeout conditions.
+
+### Changed
+- Daemon resilience test (`test_daemon_resilience.py`) hardened against race condition where `SIGKILL` test read a stale PID from a prior run that had already exited.
+
+## [0.2.0] - 2026-05-26
+
+### Added — Final Polish & Release Readiness
+- CLI usage management: `synap usage show` (displays Rich aggregated usage table and summary panel) and `synap usage clear`.
+- CLI wiki management: `synap wiki list` and `synap wiki show <filepath>` (renders page in terminal via Rich Markdown).
+- LLM call database logging: records `prompt_tokens`, `completion_tokens` for retrieval and wiki generation passes.
+- Real-time daemon state: heartbeats integrated into `synap status`, `synap doctor`, and the Web UI status endpoints.
+- Premium Web UI dashboard polish: dual L3 memory (Approved vs Pending) view, real-time LLM usage analytics, and active daemon PID badge.
+- Defensive GHA release pipeline: `.github/workflows/release.yml` automates TestPyPI and PyPI publishing, tag alignment checking, and draft release generation.
+- Clean Typer execution wrapper: intercepts configuration and credential exceptions to output actionable suggestions (e.g. `synap setup`) instead of tracebacks.
+
+### Added — Hardening Loop 2: MCP Stability & Protocol Polish
+- Deterministic MCP JSON envelope: every tool response carries `ok`, `data`, `warnings`, `trace_id`, `dirty_tree`.
+- Structured error objects with `code`, `message`, and `suggestion` fields for all failure paths.
+- `dirty_tree` propagation: agents are warned when the working tree is ahead of the index.
+- `get_approved_memory()` and `get_pending_memory()` MCP tools exposing lesson trust status.
+- `synap mcp verify` command to assert full protocol contract compliance.
+
+### Added — Hardening Loop 3: Lesson System Trust & Memory Lifecycle
+- Formal `LessonStatus` enum enforcing explicit state machine: `PENDING → APPROVED/REJECTED`, `APPROVED → EXPIRED`.
+- Retrieval gating: only `APPROVED`, non-expired lessons are injected into LLM context as `# APPROVED SYSTEM MEMORY`.
+- `prune_expired_lessons()` transitions stale lessons to `EXPIRED` state on demand.
+- `approval_actor` field on lessons for full human-governance provenance.
+- `synap memory status` — counts of pending, approved, expired lessons.
+- `synap memory prune` — forces expiry evaluation and prunes dead memory.
+- `synap memory verify` — checks approved lessons' `files_affected` against current repo state; reports dangling references.
+- `synap lessons approve <id>` and `synap lessons reject <id>` — explicit human governance over pending lessons.
+
+### Added — Hardening Loop 4: Tracing, Git Compliance & Release Hardening
+- `GitIgnoreSpec` glob-to-regex parser in `RepositoryScanner` respects `.gitignore` patterns.
+- Auto-protection: `SynapRuntime.bootstrap()` automatically adds `.synap/` to `.gitignore`.
+- Enhanced binary file detection via extension blocklist + control-character ratio analysis.
+- Symlink traversal prevention in `RepositoryScanner` (path-containment enforcement).
+- `TraceStore` writes structured operational traces to `.synap/trace_latest.json`.
+- Full hybrid retrieval tracing: latency timeline, token allocation, truncation explanations, structural hops, dirty-tree warnings.
+- `/api/v1/trace/latest` REST endpoint exposing the latest trace.
+- Diagnostic Web UI with visual latency timeline, source provenance table, and dirty-tree badge.
+- Daemon heartbeat file (`.synap/daemon_heartbeat.json`) with PID, uptime, and recovery metrics.
+- Daemon self-healing: SQLite corruption detected via `PRAGMA quick_check` triggers a wipe + re-bootstrap.
+- `synap rollback` — interactive rollback to a previous git commit with lesson preservation.
+- `synap recover` — explicit manual DB corruption recovery flow.
+- `py.typed` marker for PEP 561 compliance.
+
+### Changed
+- CI pipeline split into 4 focused jobs: `lint`, `test`, `benchmark` (main only), `release-validation`.
+- Release validation now runs `synap init` before `synap doctor` to ensure a valid Synap DB context.
+- `pytest` configured with `asyncio_default_fixture_loop_scope = "function"` to eliminate deprecation warnings.
+- Benchmark tests gated behind `benchmark` marker; skipped in fast PR CI passes.
+
+### Fixed
+- `synap doctor` in CI release validation step previously ran against an uninitialized directory.
+- `ruff format` drift in `cli/main.py` and `indexer/daemon.py` resolved.
+
+## [0.1.0] - 2026-05-25
+
+### Added
+- Core deterministic indexing engine using Tree-sitter and Git content hashes.
+- 4-stage hybrid retrieval pipeline (Temporal, Structural, Lexical, Semantic).
+- Model Context Protocol (MCP) server for IDE integration.
+- "Why-This-Context" retrieval tracing system.
+- Secure secret management via `python-keyring`.
+- `synap doctor` for system validation.
+- Diagnostic UI dashboard.
+
+### Changed
+- Refactored entire architecture from event-sourcing to deterministic Git projections.
+- Consolidated storage into unified SQLite schema with Recursive CTE support.
+- Upgraded documentation to production infrastructure standards.
+
+### Removed
+- Legacy "cognitive OS" and "graph operating system" abstractions.
+- Speculative async priority queues and replay engines.
+- Brittle regex-based parsers.

{synap_git-0.2.1 → synap_git-1.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: synap-git
-Version: 0.2.1
+Version: 1.1.1
 Summary: Persistent structural context infrastructure for AI coding agents.
 Project-URL: Homepage, https://github.com/saahilpal/synap-git
 Project-URL: Repository, https://github.com/saahilpal/synap-git
@@ -27,6 +27,7 @@ Requires-Dist: keyring==25.6.0
 Requires-Dist: markdown-it-py==3.0.0
 Requires-Dist: mcp==1.3.0
 Requires-Dist: msgpack==1.1.0
+Requires-Dist: prompt-toolkit<3.0.44
 Requires-Dist: pydantic-settings==2.8.1
 Requires-Dist: pydantic==2.10.6
 Requires-Dist: questionary==2.1.0
@@ -73,6 +74,16 @@ Synap is **NOT a RAG system.** It is a strict, deterministic background daemon t
 
 ---
 
+## ⚡ Performance Architecture (v1.1.0)
+
+Synap is built on the **Git-Snapshot Paradigm** to guarantee sub-100ms response times for everyday agent workflows:
+- **Two separate code paths:** First-run indexing uses process-pool parallel parsing (`ProcessPoolExecutor`) to build the initial codebase structure, while subsequent indexing uses a Git delta change detector (`git diff-tree`) to process only changed files. (Reason: Avoids full filesystem scans on every run).
+- **No filesystem scan / file hashing on incremental runs:** Synap uses Git blob OIDs to detect changed files. (Reason: Git already hashes everything, making filesystem reads redundant).
+- **Decoupled non-deterministic LLM pipeline:** Structural parsing (deterministic, fast) completes immediately and enqueues wiki generation (non-deterministic, slow) to a background worker, while lazy caching fallback handles CLI/API wiki requests. (Reason: Never block structural indexing or CLI commands on LLM API response latency).
+- **SQLite optimizations:** SQLite writes are grouped into a single transaction per batch (Reason: Avoids disk sync overhead per row). SQLite FTS5 index matches symbols in sub-milliseconds (Reason: Eliminates slow wildcard `LIKE` table scans). Pre-computed dot-separated `module_key` columns allow O(1) module resolution (Reason: Prevents suffix-matching scans).
+
+---
+
 ## 🏗️ High-Level Architecture (HLD)
 
 Synap bridges the gap between your local file system, Git history, and the LLM via a 3-layer indexing strategy.
@@ -227,6 +238,7 @@ Synap uses a powerful, strict CLI interface. Every destructive action prompts fo
 - `synap memory status .` : View counts of approved, pending, and expired lessons.
 - `synap memory prune .` : Prune expired rules and cleanup memory.
 - `synap memory verify .` : Detect dangling file references in active memory.
+- `synap lessons review .` : Interactively review and Approve/Edit/Reject pending lessons.
 - `synap lessons approve <id> .` : Approve a pending revert lesson to activate it.
 - `synap lessons reject <id> .` : Reject and discard a pending lesson.
 - `synap checkpoint create . --doing "..."` : Save the current context state.
@@ -236,8 +248,8 @@ Synap uses a powerful, strict CLI interface. Every destructive action prompts fo
 ### Developer Tools
 - `synap wiki list .` : List all generated wiki documentation files.
 - `synap wiki show <filepath> .` : Render a specific wiki markdown page to the console.
-- `synap cost show .` : Display detailed aggregated LLM token usage and estimated costs.
-- `synap cost clear .` : Purge all LLM call cost history.
+- `synap usage show .` : Display detailed aggregated LLM token usage.
+- `synap usage clear .` : Purge all LLM call history.
 - `synap doctor .` : Validate SQLite integrity, Tree-sitter, tokenizers, LLM providers, and daemon heartbeat.
 - `synap mcp verify .` : Verify MCP protocol, tool schemas, and contract stability.
 
@@ -251,3 +263,4 @@ Synap is built on the philosophy that AI tools must be transparent and controlla
 - Ensure all states are stored exclusively in `synap.db` or `.synap/wiki/`.
 
 License: [Apache 2.0](LICENSE.md)
+](LICENSE.md)

{synap_git-0.2.1 → synap_git-1.1.1}/README.md

@@ -24,6 +24,16 @@ Synap is **NOT a RAG system.** It is a strict, deterministic background daemon t
 
 ---
 
+## ⚡ Performance Architecture (v1.1.0)
+
+Synap is built on the **Git-Snapshot Paradigm** to guarantee sub-100ms response times for everyday agent workflows:
+- **Two separate code paths:** First-run indexing uses process-pool parallel parsing (`ProcessPoolExecutor`) to build the initial codebase structure, while subsequent indexing uses a Git delta change detector (`git diff-tree`) to process only changed files. (Reason: Avoids full filesystem scans on every run).
+- **No filesystem scan / file hashing on incremental runs:** Synap uses Git blob OIDs to detect changed files. (Reason: Git already hashes everything, making filesystem reads redundant).
+- **Decoupled non-deterministic LLM pipeline:** Structural parsing (deterministic, fast) completes immediately and enqueues wiki generation (non-deterministic, slow) to a background worker, while lazy caching fallback handles CLI/API wiki requests. (Reason: Never block structural indexing or CLI commands on LLM API response latency).
+- **SQLite optimizations:** SQLite writes are grouped into a single transaction per batch (Reason: Avoids disk sync overhead per row). SQLite FTS5 index matches symbols in sub-milliseconds (Reason: Eliminates slow wildcard `LIKE` table scans). Pre-computed dot-separated `module_key` columns allow O(1) module resolution (Reason: Prevents suffix-matching scans).
+
+---
+
 ## 🏗️ High-Level Architecture (HLD)
 
 Synap bridges the gap between your local file system, Git history, and the LLM via a 3-layer indexing strategy.
@@ -178,6 +188,7 @@ Synap uses a powerful, strict CLI interface. Every destructive action prompts fo
 - `synap memory status .` : View counts of approved, pending, and expired lessons.
 - `synap memory prune .` : Prune expired rules and cleanup memory.
 - `synap memory verify .` : Detect dangling file references in active memory.
+- `synap lessons review .` : Interactively review and Approve/Edit/Reject pending lessons.
 - `synap lessons approve <id> .` : Approve a pending revert lesson to activate it.
 - `synap lessons reject <id> .` : Reject and discard a pending lesson.
 - `synap checkpoint create . --doing "..."` : Save the current context state.
@@ -187,8 +198,8 @@ Synap uses a powerful, strict CLI interface. Every destructive action prompts fo
 ### Developer Tools
 - `synap wiki list .` : List all generated wiki documentation files.
 - `synap wiki show <filepath> .` : Render a specific wiki markdown page to the console.
-- `synap cost show .` : Display detailed aggregated LLM token usage and estimated costs.
-- `synap cost clear .` : Purge all LLM call cost history.
+- `synap usage show .` : Display detailed aggregated LLM token usage.
+- `synap usage clear .` : Purge all LLM call history.
 - `synap doctor .` : Validate SQLite integrity, Tree-sitter, tokenizers, LLM providers, and daemon heartbeat.
 - `synap mcp verify .` : Verify MCP protocol, tool schemas, and contract stability.
 
@@ -202,3 +213,4 @@ Synap is built on the philosophy that AI tools must be transparent and controlla
 - Ensure all states are stored exclusively in `synap.db` or `.synap/wiki/`.
 
 License: [Apache 2.0](LICENSE.md)
+](LICENSE.md)

{synap_git-0.2.1 → synap_git-1.1.1}/pyproject.toml

@@ -27,6 +27,7 @@ dependencies = [
   "markdown-it-py==3.0.0",
   "mcp==1.3.0",
   "msgpack==1.1.0",
+  "prompt-toolkit<3.0.44",
   "pydantic==2.10.6",
   "pydantic-settings==2.8.1",
   "questionary==2.1.0",

{synap_git-0.2.1 → synap_git-1.1.1}/src/synap_git/__init__.py

@@ -2,4 +2,4 @@
 
 __all__ = ["__version__"]
 
-__version__ = "0.2.1"
+__version__ = "1.1.1"

{synap_git-0.2.1 → synap_git-1.1.1}/src/synap_git/api/app.py

@@ -10,6 +10,7 @@ from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import HTMLResponse, StreamingResponse
 from fastapi.staticfiles import StaticFiles
 
+from synap_git import __version__
 from synap_git.indexer.engine import SynapRuntime
 
 
@@ -17,7 +18,7 @@ def create_app(runtime: SynapRuntime) -> FastAPI:
     app = FastAPI(
         title="Synap Context Diagnostics",
         description="Deterministic structural context infrastructure for AI coding agents.",
-        version="0.2.0",
+        version=__version__,
     )
 
     app.add_middleware(
@@ -36,17 +37,20 @@ def create_app(runtime: SynapRuntime) -> FastAPI:
     @app.get("/", response_class=HTMLResponse)
     async def get_index() -> str:
         index_file = static_dir / "index.html"
-        if index_file.exists():
-            return index_file.read_text(encoding="utf-8")
+        exists = await asyncio.to_thread(index_file.exists)
+        if exists:
+            return await asyncio.to_thread(index_file.read_text, encoding="utf-8")
         return "<h1>Synap Diagnostic UI is ready.</h1>"
 
     @app.get("/api/v1/status")
     async def get_status() -> dict[str, Any]:
         try:
-            status = runtime.status()
+            status = await asyncio.to_thread(runtime.status)
             from synap_git.cli.main import _read_daemon_heartbeat
 
-            daemon_info = _read_daemon_heartbeat(Path(status.repository_path))
+            daemon_info = await asyncio.to_thread(
+                _read_daemon_heartbeat, Path(status.repository_path)
+            )
             return {
                 "repository_path": status.repository_path,
                 "branch": status.branch,
@@ -63,7 +67,9 @@ def create_app(runtime: SynapRuntime) -> FastAPI:
     @app.get("/api/v1/trace/latest")
     async def get_latest_trace() -> dict[str, Any]:
         try:
-            return runtime.trace_store.get_latest()
+            if runtime.trace_store:
+                return await asyncio.to_thread(runtime.trace_store.get_latest)
+            return {}
         except Exception as exc:
             raise HTTPException(status_code=500, detail=str(exc))
 
@@ -73,7 +79,7 @@ def create_app(runtime: SynapRuntime) -> FastAPI:
 
         async def event_generator() -> AsyncGenerator[str, None]:
             while True:
-                status = runtime.status()
+                status = await asyncio.to_thread(runtime.status)
                 yield f"data: {json.dumps(status.__dict__)}\n\n"
                 await asyncio.sleep(2)
 
@@ -81,31 +87,56 @@ def create_app(runtime: SynapRuntime) -> FastAPI:
 
     @app.get("/wiki/{filepath:path}")
     async def get_wiki_page(filepath: str) -> dict[str, Any]:
-        wiki_path = runtime.wiki.wiki_dir / f"{filepath}.md"
-        if wiki_path.exists():
-            return {"status": "ok", "content": wiki_path.read_text(encoding="utf-8")}
+        base_dir = runtime.wiki.wiki_dir.resolve()
+        try:
+            target_path = (base_dir / filepath).resolve()
+        except Exception:
+            raise HTTPException(status_code=400, detail="Invalid path")
+
+        if not target_path.is_relative_to(base_dir):
+            raise HTTPException(status_code=403, detail="Path traversal detected")
+
+        target = filepath
+        if target.endswith(".md"):
+            target = target[:-3]
+        try:
+            await asyncio.to_thread(runtime.wiki.ensure_wiki_page, target)
+        except Exception:
+            pass
+
+        wiki_path = runtime.wiki.wiki_dir / f"{target}.md"
+        exists = await asyncio.to_thread(wiki_path.exists)
+        if exists:
+            content = await asyncio.to_thread(wiki_path.read_text, encoding="utf-8")
+            return {"status": "ok", "content": content}
         return {"status": "error", "message": "Wiki not found"}
 
     @app.get("/api/v1/memory")
     async def get_memory_page() -> dict[str, Any]:
-        approved = runtime.store.get_lessons("approved")
-        pending = runtime.store.get_lessons("pending")
+        approved = await asyncio.to_thread(runtime.store.get_lessons, "approved")
+        pending = await asyncio.to_thread(runtime.store.get_lessons, "pending")
         return {"status": "ok", "approved": approved, "pending": pending}
 
-    @app.get("/api/v1/cost")
-    async def get_cost_page() -> dict[str, Any]:
+    def _fetch_calls() -> list[dict[str, Any]]:
         with runtime.store.connect() as conn:
             rows = conn.execute("SELECT * FROM llm_calls ORDER BY created_at DESC").fetchall()
-            calls = [dict(r) for r in rows]
+            return [dict(r) for r in rows]
+
+    @app.get("/api/v1/usage")
+    async def get_usage_page() -> dict[str, Any]:
+        calls = await asyncio.to_thread(_fetch_calls)
         return {"status": "ok", "calls": calls}
 
-    @app.get("/api/v1/checkpoints")
-    async def get_checkpoints_page() -> dict[str, Any]:
+    def _fetch_checkpoints() -> list[dict[str, Any]]:
         with runtime.store.connect() as conn:
             rows = conn.execute(
                 "SELECT * FROM checkpoints ORDER BY created_at DESC LIMIT 20"
             ).fetchall()
-            cps = [dict(r) for r in rows]
+            return [dict(r) for r in rows]
+
+    @app.get("/api/v1/checkpoints")
+    async def get_checkpoints_page() -> dict[str, Any]:
+        cps = await asyncio.to_thread(_fetch_checkpoints)
         return {"status": "ok", "checkpoints": cps}
 
     return app

{synap_git-0.2.1 → synap_git-1.1.1}/src/synap_git/api/static/index.html

@@ -202,15 +202,11 @@
             </div>
         </div>
 
-        <!-- LLM Cost & Usage -->
+        <!-- LLM Usage -->
         <div class="card" style="grid-column: span 2;">
-            <h2>LLM Call Cost & Usage</h2>
+            <h2>LLM Call Usage</h2>
             <div style="display: grid; grid-template-columns: 1fr 2fr; gap: 1.5rem;">
                 <div class="stats-grid" style="grid-template-columns: 1fr; gap: 0.75rem;">
-                    <div class="stat-box">
-                        <div class="stat-value" id="cost-total-usd">$0.000000</div>
-                        <div class="stat-label">Total Cost (USD)</div>
-                    </div>
                     <div class="stat-box">
                         <div class="stat-value" id="cost-total-calls">0</div>
                         <div class="stat-label">Total LLM Calls</div>
@@ -231,11 +227,10 @@
                                     <th>Model</th>
                                     <th>Purpose</th>
                                     <th>Tokens</th>
-                                    <th>Cost</th>
                                 </tr>
                             </thead>
                             <tbody id="cost-calls-body">
-                                <tr><td colspan="6" class="empty-state">No calls recorded.</td></tr>
+                                <tr><td colspan="5" class="empty-state">No calls recorded.</td></tr>
                             </tbody>
                         </table>
                     </div>
@@ -443,44 +438,59 @@
                     pendingBody.innerHTML = `<tr><td colspan="3" class="empty-state">No pending lessons needing approval.</td></tr>`;
                 }
 
-                // Fetch Cost & Usage
-                const costRes = await fetch('/api/v1/cost');
-                const costData = await costRes.json();
+                // Fetch Usage
+                const usageRes = await fetch('/api/v1/usage');
+                const usageData = await usageRes.json();
                 const costBody = document.getElementById('cost-calls-body');
 
-                let totalCost = 0;
                 let totalCalls = 0;
                 let totalTokens = 0;
 
-                if (costData.calls && costData.calls.length > 0) {
-                    totalCalls = costData.calls.length;
-                    costBody.innerHTML = costData.calls.slice(0, 15).map(c => {
-                        totalCost += c.cost_usd;
+                if (usageData.calls && usageData.calls.length > 0) {
+                    totalCalls = usageData.calls.length;
+                    usageData.calls.slice(0, 15).map(c => {
                         const tokens = c.input_tokens + c.output_tokens;
                         totalTokens += tokens;
 
                         const timeStr = new Date(c.created_at * 1000).toLocaleTimeString();
-                        return `
+                        const row = `
                         <tr class="row-hover">
                             <td>${timeStr}</td>
                             <td>${c.provider}</td>
                             <td><code>${c.model}</code></td>
                             <td>${c.purpose}</td>
                             <td>${tokens.toLocaleString()}</td>
-                            <td style="color:var(--success); font-family:var(--font-mono); font-size:11px;">$${c.cost_usd.toFixed(6)}</td>
                         </tr>
                         `;
-                    }).join('');
+                        costBody.insertAdjacentHTML('beforeend', row);
+                    });
+
+                    // Clear the loading message if it exists
+                    if (usageData.calls.length > 0 && costBody.querySelector('.empty-state')) {
+                        costBody.innerHTML = '';
+                        // Re-render the 15 calls because we just cleared it
+                         costBody.innerHTML = usageData.calls.slice(0, 15).map(c => {
+                            const tokens = c.input_tokens + c.output_tokens;
+                            const timeStr = new Date(c.created_at * 1000).toLocaleTimeString();
+                            return `
+                            <tr class="row-hover">
+                                <td>${timeStr}</td>
+                                <td>${c.provider}</td>
+                                <td><code>${c.model}</code></td>
+                                <td>${c.purpose}</td>
+                                <td>${tokens.toLocaleString()}</td>
+                            </tr>
+                            `;
+                        }).join('');
+                    }
 
-                    costData.calls.slice(15).forEach(c => {
-                        totalCost += c.cost_usd;
+                    usageData.calls.slice(15).forEach(c => {
                         totalTokens += (c.input_tokens + c.output_tokens);
                     });
                 } else {
-                    costBody.innerHTML = `<tr><td colspan="6" class="empty-state">No LLM calls recorded yet.</td></tr>`;
+                    costBody.innerHTML = `<tr><td colspan="5" class="empty-state">No LLM calls recorded yet.</td></tr>`;
                 }
 
-                document.getElementById('cost-total-usd').textContent = `$${totalCost.toFixed(6)}`;
                 document.getElementById('cost-total-calls').textContent = totalCalls.toLocaleString();
                 document.getElementById('cost-total-tokens').textContent = totalTokens.toLocaleString();