@evomap/evolver 1.75.0 → 1.77.0

package/README.ja-JP.md

@@ -180,6 +180,8 @@ evolver --loop
 | ループ (`evolver --loop`) | 適応的スリープ付きのデーモンループで上記を繰り返す |
 | OpenClaw 内 | ホストランタイムが `sessions_spawn(...)` などの stdout ディレクティブを解釈 |
 
+> **`--loop` は「動作中のエージェントをリアルタイムで支援する」モードではありません。** ループモードはバックグラウンドの自己メンテナンス（validator 実行、worker タスク、ATP マーチャント自動配信、solidify）のためのもので、その stdout は evolver 自身が消費します。したがって、たとえ OpenClaw / Cursor / Claude Code がインストールされていても、ループモードで出力される `sessions_spawn(...)` ディレクティブはこれらのホストには届きません。evolver にライブセッションを観察・補助させたい場合は、そのエージェントセッションの **内部から** `evolver run` を呼び出してください（OpenClaw はその単一ランの stdout ディレクティブを取り込みます）。OpenClaw ユーザーはさらに、`AGENT_NAME`（または `AGENT_SESSIONS_DIR`）が実際にセッションを生成しているエージェントのディレクトリ（`~/.openclaw/agents/<名前>/sessions/`）を指していることを確認してください -- さもないと evolver は自身のログにフォールバックし、「空転している」ように見えます。
+
 ## 対象ユーザー
 
 **向いている**

package/README.ko-KR.md

@@ -179,6 +179,8 @@ evolver --loop
 | 루프 (`evolver --loop`) | 적응형 슬립이 포함된 데몬 루프에서 위 과정을 반복 |
 | OpenClaw 내부 | 호스트 런타임이 `sessions_spawn(...)` 등 stdout 지시문을 해석 |
 
+> **`--loop`은 "실행 중인 에이전트를 실시간으로 보조하는" 모드가 아닙니다.** 루프 모드는 백그라운드 자가 유지보수(validator 실행, worker 작업, ATP 상인 자동 배달, solidify)를 위한 것이며, 그 stdout은 evolver 자신이 소비합니다. 따라서 OpenClaw / Cursor / Claude Code가 설치되어 있더라도, 루프 모드에서 출력되는 `sessions_spawn(...)` 지시문은 이 호스트들에 전달되지 않습니다. 라이브 세션을 evolver가 관찰·보조하게 하려면, 해당 에이전트 세션 **내부에서** `evolver run`을 호출하세요(OpenClaw는 그 단일 실행의 stdout 지시문을 처리합니다). OpenClaw 사용자는 추가로, `AGENT_NAME`(또는 `AGENT_SESSIONS_DIR`)이 실제로 세션을 생성하는 에이전트 디렉터리(`~/.openclaw/agents/<이름>/sessions/`)를 가리키는지 확인하세요 -- 그렇지 않으면 evolver는 자신의 로그로 폴백하며, "빈 사이클만 돌고 있는" 것처럼 보입니다.
+
 ## 대상 사용자
 
 **적합한 경우**

package/README.md

@@ -178,6 +178,8 @@ When running inside a host runtime (e.g., [OpenClaw](https://openclaw.com)), the
 | Loop (`evolver --loop`) | Repeats the above in a daemon loop with adaptive sleep |
 | Inside OpenClaw | Host runtime interprets stdout directives like `sessions_spawn(...)` |
 
+> **`--loop` is not a real-time agent assistant.** Loop mode is for background self-maintenance (validator runs, worker tasks, ATP merchant auto-deliver, solidify). Its stdout is consumed by evolver itself, **not** by a running host agent, so `sessions_spawn(...)` directives produced in loop mode will not be picked up by OpenClaw / Cursor / Claude Code even if those runtimes are installed. If you want evolver to observe and advise a live agent session, call `evolver run` from **inside** that agent session (OpenClaw will pick up the stdout directives on that single run). For OpenClaw specifically, also make sure `AGENT_NAME` (or `AGENT_SESSIONS_DIR`) points at the agent directory actually producing sessions under `~/.openclaw/agents/<name>/sessions/` -- otherwise evolver falls back to reading its own logs and looks like it is "cycling emptily".
+
 ## Who This Is For / Not For
 
 **For**

package/README.zh-CN.md

@@ -177,6 +177,8 @@ evolver --loop
 | 循环模式 (`evolver --loop`) | 在守护进程循环中重复上述流程，带自适应休眠 |
 | 在 OpenClaw 中 | 宿主运行时解释 stdout 中的指令（如 `sessions_spawn(...)`） |
 
+> **`--loop` 不是"实时辅助正在干活的 agent"的模式。** 循环模式用于后台自维护任务（validator 验证、worker 任务、ATP 商家自动交付、solidify），它的 stdout 是被 evolver 自己消费的，**不会**传给正在运行的 OpenClaw / Cursor / Claude Code agent——即使这些宿主已经安装，`sessions_spawn(...)` 指令在循环模式下也不会被它们接收。如果你想让 evolver 观察并辅助一次具体的 agent 会话，请在那个 agent 会话内部调用 `evolver run`（一次一轮），OpenClaw 会在这次运行中接管 stdout 指令。对 OpenClaw 用户还要特别注意：`AGENT_NAME`（或 `AGENT_SESSIONS_DIR`）必须指向真正在产生 session 的那个 agent 目录（`~/.openclaw/agents/<名字>/sessions/`），否则 evolver 会回退到读自己的日志，看上去就像在"空转"。
+
 ## 适用 / 不适用场景
 
 **适用**

package/assets/gep/candidates.jsonl

@@ -1,3 +1,2 @@
-{"type":"CapabilityCandidate","id":"cand_b9a66a5c","title":"Harden session log detection and fallback behavior","source":"signals","created_at":"2026-04-28T12:53:54.871Z","signals":["memory_missing","user_missing","session_logs_missing"],"tags":["memory_missing","user_missing","session_logs_missing","area:memory"],"shape":{"title":"Harden session log detection and fallback behavior","input":"Recent session transcript + memory snippets + user instructions","output":"A safe, auditable evolution patch guided by GEP assets","invariants":"Protocol order, small reversible patches, validation, append-only events","params":"Signals: memory_missing, user_missing, session_logs_missing","failure_points":"Missing signals, over-broad changes, skipped validation, missing knowledge solidification","evidence":"Signal present: session_logs_missing"}}
-{"type":"CapabilityCandidate","id":"cand_b9a66a5c","title":"Harden session log detection and fallback behavior","source":"signals","created_at":"2026-04-28T12:54:19.044Z","signals":["memory_missing","user_missing","session_logs_missing"],"tags":["memory_missing","user_missing","session_logs_missing","area:memory"],"shape":{"title":"Harden session log detection and fallback behavior","input":"Recent session transcript + memory snippets + user instructions","output":"A safe, auditable evolution patch guided by GEP assets","invariants":"Protocol order, small reversible patches, validation, append-only events","params":"Signals: memory_missing, user_missing, session_logs_missing","failure_points":"Missing signals, over-broad changes, skipped validation, missing knowledge solidification","evidence":"Signal present: session_logs_missing"}}
-{"type":"CapabilityCandidate","id":"cand_b9a66a5c","title":"Harden session log detection and fallback behavior","source":"signals","created_at":"2026-04-28T12:54:24.039Z","signals":["memory_missing","user_missing","session_logs_missing"],"tags":["memory_missing","user_missing","session_logs_missing","area:memory"],"shape":{"title":"Harden session log detection and fallback behavior","input":"Recent session transcript + memory snippets + user instructions","output":"A safe, auditable evolution patch guided by GEP assets","invariants":"Protocol order, small reversible patches, validation, append-only events","params":"Signals: memory_missing, user_missing, session_logs_missing","failure_points":"Missing signals, over-broad changes, skipped validation, missing knowledge solidification","evidence":"Signal present: session_logs_missing"}}
+{"type":"CapabilityCandidate","id":"cand_b9a66a5c","title":"Harden session log detection and fallback behavior","source":"signals","created_at":"2026-04-30T13:31:51.491Z","signals":["memory_missing","user_missing","session_logs_missing"],"tags":["memory_missing","user_missing","session_logs_missing","area:memory"],"shape":{"title":"Harden session log detection and fallback behavior","input":"Recent session transcript + memory snippets + user instructions","output":"A safe, auditable evolution patch guided by GEP assets","invariants":"Protocol order, small reversible patches, validation, append-only events","params":"Signals: memory_missing, user_missing, session_logs_missing","failure_points":"Missing signals, over-broad changes, skipped validation, missing knowledge solidification","evidence":"Signal present: session_logs_missing"}}
+{"type":"CapabilityCandidate","id":"cand_b9a66a5c","title":"Harden session log detection and fallback behavior","source":"signals","created_at":"2026-04-30T13:31:54.079Z","signals":["memory_missing","user_missing","session_logs_missing"],"tags":["memory_missing","user_missing","session_logs_missing","area:memory"],"shape":{"title":"Harden session log detection and fallback behavior","input":"Recent session transcript + memory snippets + user instructions","output":"A safe, auditable evolution patch guided by GEP assets","invariants":"Protocol order, small reversible patches, validation, append-only events","params":"Signals: memory_missing, user_missing, session_logs_missing","failure_points":"Missing signals, over-broad changes, skipped validation, missing knowledge solidification","evidence":"Signal present: session_logs_missing"}}

package/index.js

@@ -194,6 +194,35 @@ async function main() {
         }
         console.log(`Loop mode enabled (internal daemon, bridge=${process.env.EVOLVE_BRIDGE}, verbose=${isVerbose}).`);
 
+        // Startup diagnostic: in daemon mode evolver consumes its own stdout
+        // instead of handing `sessions_spawn(...)` directives to a host
+        // runtime (OpenClaw). If the operator expects real-time agent assist
+        // they are likely using the wrong mode; if they intend daemon mode
+        // they still need AGENT_NAME / AGENT_SESSIONS_DIR pointing at a live
+        // agent or the loop will just cycle on its own logs. Emit a single
+        // warning at startup so "empty cycling" has a visible breadcrumb.
+        try {
+          const { diagnoseSessionSourceEmpty } = require('./src/evolve');
+          const diag = diagnoseSessionSourceEmpty();
+          const hasAnySource = diag.agentSessionsDirExists ||
+            diag.cursorDirExists || diag.claudeDirExists || diag.codexDirExists ||
+            Boolean(diag.cursorTranscriptsDir);
+          if (!hasAnySource) {
+            console.warn('[Daemon] No session sources detected at startup. Loop mode runs background self-maintenance but cannot observe a live agent without at least one of:');
+            console.warn(`  - ~/.openclaw/agents/<AGENT_NAME>/sessions/ (current AGENT_NAME=${diag.agentName}, exists=${diag.agentSessionsDirExists})`);
+            console.warn('  - ~/.cursor / ~/.claude / ~/.codex (IDE transcripts)');
+            console.warn('  - EVOLVER_CURSOR_TRANSCRIPTS_DIR (explicit override)');
+            if (diag.availableOpenClawAgents.length > 0) {
+              console.warn(`  Available OpenClaw agents under ~/.openclaw/agents/: ${diag.availableOpenClawAgents.join(', ')}`);
+              console.warn('  Set AGENT_NAME=<agent> or AGENT_SESSIONS_DIR=<abs path> to the one actually doing work.');
+            }
+            for (const hint of diag.hints) {
+              console.warn(`  HINT: ${hint}`);
+            }
+            console.warn('  If you want real-time agent assist (not background self-maintenance), run `evolver run` from inside the agent session instead of `evolver --loop`.');
+          }
+        } catch (_diagErr) { /* diagnostics must never block startup */ }
+
         const { getEvolutionDir, getEvolverLogPath } = require('./src/gep/paths');
         const solidifyStatePath = path.join(getEvolutionDir(), 'evolution_solidify_state.json');
 
@@ -996,6 +1025,175 @@ async function main() {
       process.exit(1);
     }
 
+  } else if (command === 'sync') {
+    const { getHubUrl, getNodeId, buildHubHeaders, sendHelloToHub, getHubNodeSecret } = require('./src/gep/a2aProtocol');
+    const { upsertGene, upsertCapsule, loadGenes, loadCapsules } = require('./src/gep/assetStore');
+
+    const hubUrl = getHubUrl();
+    if (!hubUrl) {
+      console.error('[sync] A2A_HUB_URL is not configured.');
+      process.exit(1);
+    }
+
+    try {
+      if (!getHubNodeSecret()) {
+        console.log('[sync] No node_secret found. Sending hello to Hub to register...');
+        const helloResult = await sendHelloToHub();
+        if (!helloResult || !helloResult.ok) {
+          console.error('[sync] Failed to register with Hub:', helloResult && helloResult.error || 'unknown');
+          process.exit(1);
+        }
+        console.log('[sync] Registered as ' + getNodeId());
+      }
+
+      const nodeId = getNodeId();
+      const baseUrl = hubUrl.replace(/\/+$/, '');
+      const typeFilter = (function () {
+        const f = args.find(function (a) { return typeof a === 'string' && a.startsWith('--type='); });
+        return f ? f.slice('--type='.length) : null;
+      })();
+      const dryRun = args.includes('--dry-run');
+      const limitPerPage = 100;
+
+      console.log('[sync] Fetching purchased assets from Hub...');
+      let allAssets = [];
+      let cursor = null;
+      let page = 0;
+
+      while (true) {
+        page++;
+        let url = baseUrl + '/a2a/assets/purchased?node_id=' + encodeURIComponent(nodeId) + '&limit=' + limitPerPage;
+        if (cursor) url += '&cursor=' + encodeURIComponent(cursor);
+        if (typeFilter) url += '&type=' + encodeURIComponent(typeFilter);
+
+        const resp = await fetch(url, {
+          method: 'GET',
+          headers: buildHubHeaders(),
+          signal: AbortSignal.timeout(30000),
+        });
+
+        if (!resp.ok) {
+          const body = await resp.text().catch(function () { return ''; });
+          console.error('[sync] Hub returned HTTP ' + resp.status + ': ' + body.slice(0, 500));
+          process.exit(1);
+        }
+
+        const data = await resp.json();
+        if (Array.isArray(data.assets)) {
+          allAssets = allAssets.concat(data.assets);
+        }
+        if (isVerbose) console.log('[sync] Page ' + page + ': ' + (data.count || 0) + ' assets (total so far: ' + allAssets.length + ')');
+
+        if (data.has_more && data.next_cursor) {
+          cursor = data.next_cursor;
+        } else {
+          break;
+        }
+      }
+
+      if (allAssets.length === 0) {
+        console.log('[sync] No purchased assets found on Hub.');
+        process.exit(0);
+      }
+
+      console.log('[sync] Found ' + allAssets.length + ' purchased asset(s). Syncing to local store...');
+
+      const existingGenes = loadGenes();
+      const existingCapsules = loadCapsules();
+      const localGeneIds = new Set(existingGenes.filter(function (g) { return g && g.id; }).map(function (g) { return g.id; }));
+      const localCapsuleIds = new Set(existingCapsules.filter(function (c) { return c && c.id; }).map(function (c) { return c.id; }));
+
+      let synced = 0;
+      let skipped = 0;
+      let fetchErrors = 0;
+
+      for (const asset of allAssets) {
+        const assetId = asset.asset_id;
+        const assetType = asset.asset_type;
+
+        if (assetType === 'Gene' && localGeneIds.has(asset.local_id || assetId)) {
+          skipped++;
+          continue;
+        }
+        if (assetType === 'Capsule' && localCapsuleIds.has(asset.local_id || assetId)) {
+          skipped++;
+          continue;
+        }
+        if (assetType !== 'Gene' && assetType !== 'Capsule') {
+          skipped++;
+          continue;
+        }
+
+        if (dryRun) {
+          console.log('  [dry-run] Would sync: ' + assetType + ' ' + assetId);
+          synced++;
+          continue;
+        }
+
+        try {
+          const detailResp = await fetch(baseUrl + '/a2a/assets/' + encodeURIComponent(assetId) + '?detailed=true', {
+            method: 'GET',
+            headers: buildHubHeaders(),
+            signal: AbortSignal.timeout(15000),
+          });
+
+          if (!detailResp.ok) {
+            if (isVerbose) console.warn('  [sync] Failed to fetch detail for ' + assetId + ' (HTTP ' + detailResp.status + ')');
+            fetchErrors++;
+            continue;
+          }
+
+          const detail = await detailResp.json();
+          const payload = detail.payload || {};
+
+          if (assetType === 'Gene') {
+            const geneObj = {
+              type: 'Gene',
+              id: payload.id || asset.local_id || assetId,
+              category: payload.category || 'unknown',
+              signals: Array.isArray(payload.signals) ? payload.signals : [],
+              strategy: Array.isArray(payload.strategy) ? payload.strategy : [],
+              avoid: Array.isArray(payload.avoid) ? payload.avoid : [],
+              validation: payload.validation || {},
+              summary: payload.summary || detail.summary || asset.summary || '',
+              hub_asset_id: assetId,
+              synced_at: new Date().toISOString(),
+            };
+            upsertGene(geneObj);
+            localGeneIds.add(geneObj.id);
+          } else {
+            const capsuleObj = {
+              type: 'Capsule',
+              id: payload.id || asset.local_id || assetId,
+              gene: payload.gene || null,
+              genes_used: Array.isArray(payload.genes_used) ? payload.genes_used : [],
+              outcome: payload.outcome || {},
+              execution_trace: payload.execution_trace || {},
+              summary: payload.summary || detail.summary || asset.summary || '',
+              hub_asset_id: assetId,
+              synced_at: new Date().toISOString(),
+            };
+            upsertCapsule(capsuleObj);
+            localCapsuleIds.add(capsuleObj.id);
+          }
+          synced++;
+        } catch (fetchErr) {
+          if (isVerbose) console.warn('  [sync] Error fetching ' + assetId + ': ' + (fetchErr && fetchErr.message || fetchErr));
+          fetchErrors++;
+        }
+      }
+
+      console.log('[sync] Done. synced=' + synced + ' skipped=' + skipped + ' errors=' + fetchErrors);
+      if (dryRun) console.log('[sync] (dry-run mode: no files were modified)');
+    } catch (error) {
+      if (error && error.name === 'TimeoutError') {
+        console.error('[sync] Request timed out. Check your network and A2A_HUB_URL.');
+      } else {
+        console.error('[sync] Error: ' + (error && error.message || error));
+      }
+      process.exit(1);
+    }
+
   } else if (command === 'asset-log') {
     const { summarizeCallLog, readCallLog, getLogPath } = require('./src/gep/assetCallLog');
 
@@ -1140,10 +1338,13 @@ async function main() {
     }
 
   } else {
-    console.log(`Usage: node index.js [run|/evolve|solidify|review|distill|fetch|asset-log|setup-hooks|buy|orders|verify|atp-complete] [--loop]
+    console.log(`Usage: node index.js [run|/evolve|solidify|review|distill|fetch|sync|asset-log|setup-hooks|buy|orders|verify|atp-complete] [--loop]
   - fetch flags:
     - --skill=<id> | -s <id>   (skill ID to download)
     - --out=<dir>              (output directory, default: ./skills/<skill_id>)
+  - sync flags:
+    - --type=Gene|Capsule      (filter by asset type)
+    - --dry-run                (preview without writing to local store)
   - solidify flags:
     - --dry-run
     - --no-rollback

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@evomap/evolver",
-  "version": "1.75.0",
+  "version": "1.77.0",
   "description": "A GEP-powered self-evolution engine for AI agents. Features automated log analysis and Genome Evolution Protocol (GEP) for auditable, reusable evolution assets.",
   "main": "index.js",
   "bin": {