@remixmate/cli 0.9.13 → 0.9.15

package/skills/gen-digital-human/skill.json

@@ -5,6 +5,7 @@
   "category": "asset",
   "title": "Digital-Human Talking-Head",
   "description": "Digital-human video: list available avatars; produce a talking-head video from text via TTS, or drive an avatar from an existing audio URL.",
+  "auth": "required",
   "envVars": ["PRIV_TOKEN", "MM_API_BASE_URL", "AGENT_NAME"],
   "entry": { "type": "http", "handler": "gen-digital-human" },
   "parameters": {

package/skills/gen-image/SKILL.md

@@ -122,6 +122,18 @@ remixmate gen-image \
 | `--api-base-url` | Override API root | see above |
 | `--priv-token` | Override token | see above |
 
+## Credits
+
+Every run charges credits. The CLI prints a footer on stdout when it does:
+
+```
+💳 Charged 31 credits · balance 1,240
+```
+
+Relay it to the user whenever it appears — it is the only signal they get about what a
+generation cost, and the balance is the only warning before a run fails with
+`insufficient_credits`. Do not drop it from your summary.
+
 ## Error handling
 
 - **401** / **token missing** (non-OpenClaw): set `PRIV_TOKEN`.

package/skills/gen-image/skill.json

@@ -5,6 +5,7 @@
   "category": "asset",
   "title": "AI Image Generation",
   "description": "AI image generation: produce an image from a text prompt. Supports Seedream and Gemini models, plus image-to-image with reference images.",
+  "auth": "required",
   "envVars": ["PRIV_TOKEN", "MM_API_BASE_URL", "AGENT_NAME", "MM_IMAGE_MODEL"],
   "entry": { "type": "http", "handler": "gen-image" },
   "parameters": {

package/skills/gen-script/skill.json

@@ -5,6 +5,7 @@
   "category": "authoring",
   "title": "Video Script Generation",
   "description": "Video script generation: turn a topic into a structured Video DSL (JSON) that describes the full video — scene structure, asset requirements, and narrative flow.",
+  "auth": "none",
   "envVars": ["DEFAULT_IMAGE_MODEL", "DEFAULT_VIDEO_MODEL", "STUB_IMAGE_URL", "STUB_VIDEO_URL"],
   "scriptPath": "scripts/gen_script.py",
   "parameters": {

package/skills/gen-video/SKILL.md

@@ -150,6 +150,18 @@ remixmate gen-video \
 - Describe motion explicitly.
 - Example: `"At sunrise, an aerial drone shot of a futuristic city, golden light on glass facades, mist swirling, 4K ultra-clear"`.
 
+## Credits
+
+Every run charges credits. The CLI prints a footer on stdout when it does:
+
+```
+💳 Charged 31 credits · balance 1,240
+```
+
+Relay it to the user whenever it appears — it is the only signal they get about what a
+generation cost, and the balance is the only warning before a run fails with
+`insufficient_credits`. Do not drop it from your summary.
+
 ## Error handling
 
 - **401** / **token missing** (non-OpenClaw): set `PRIV_TOKEN`.

package/skills/gen-video/skill.json

@@ -5,6 +5,7 @@
   "category": "asset",
   "title": "AI Video Generation",
   "description": "AI video generation: produce a short video clip from a text prompt. Supports Seedance and Veo models, plus first/last frame and reference images.",
+  "auth": "required",
   "envVars": ["PRIV_TOKEN", "MM_API_BASE_URL", "AGENT_NAME", "MM_VIDEO_MODEL"],
   "entry": { "type": "http", "handler": "gen-video" },
   "parameters": {

package/skills/gen-voice/SKILL.md

@@ -95,6 +95,18 @@ remixmate gen-voice --text "<text-to-synthesize>" --json-output
 | Voice-over / narration | 1.0–1.2 |
 | Fast announcement | 1.2–1.5 |
 
+## Credits
+
+Every run charges credits. The CLI prints a footer on stdout when it does:
+
+```
+💳 Charged 31 credits · balance 1,240
+```
+
+Relay it to the user whenever it appears — it is the only signal they get about what a
+generation cost, and the balance is the only warning before a run fails with
+`insufficient_credits`. Do not drop it from your summary.
+
 ## Error handling
 
 - **401** / **token missing** (non-OpenClaw): set `PRIV_TOKEN`.

package/skills/gen-voice/skill.json

@@ -5,6 +5,7 @@
   "category": "asset",
   "title": "Text-to-Speech (Minimax)",
   "description": "Text-to-speech (TTS): synthesize narration audio from text via the Minimax TTS model. Returns the persisted audio URL — no download needed.",
+  "auth": "required",
   "envVars": ["PRIV_TOKEN", "MM_API_BASE_URL", "AGENT_NAME"],
   "entry": { "type": "http", "handler": "gen-voice" },
   "parameters": {

package/skills/prepare-video-assets/SKILL.md

@@ -195,6 +195,18 @@ The Asset Resolver handles each `AssetRef` in this order:
 4. **Parallel generation**: assets of the same type are generated in parallel; different types are sequenced by dependency.
 5. **Cache reuse**: assets with the same payload are checked against `asset-cache-dir` to avoid duplicate generation.
 
+## Credits
+
+Every run charges credits. The CLI prints a footer on stdout when it does:
+
+```
+💳 Charged 31 credits · balance 1,240
+```
+
+Relay it to the user whenever it appears — it is the only signal they get about what a
+generation cost, and the balance is the only warning before a run fails with
+`insufficient_credits`. Do not drop it from your summary.
+
 ## Error handling
 
 - **DSL validation failed**: pre-validate with `gen-script --validate`.

package/skills/prepare-video-assets/skill.json

@@ -5,6 +5,7 @@
   "category": "authoring",
   "title": "Video Asset Preparation",
   "description": "Resolves and generates every asset (image / audio / video) referenced by a Video DSL, persists a RenderPlan to the database, and returns a job_id for the subsequent render_video call. This is Phase 1 of the two-phase video pipeline; Phase 3 (Remotion render) lives in render_video.",
+  "auth": "required",
   "envVars": ["PRIV_TOKEN", "MM_API_BASE_URL", "MM_BACKEND_API_URL", "AGENT_NAME", "REMOTION_RENDER_API_URL", "REMOTION_RENDER_MODE", "REMOTION_OUTPUT_DIR", "ASSET_CACHE_DIR"],
   "scriptPath": "scripts/prepare_video_assets.py",
   "parameters": {

package/skills/render-video/SKILL.md

@@ -210,6 +210,18 @@ The Remotion render project lives at the monorepo root under `<monorepo-root>/re
 - Production (default): the client uses `https://api-render.remixmate.com` when `REMOTION_RENDER_API_URL` is unset — no config needed.
 - Local debugging: in the renderer directory run `npm install && npm run server:dev`, then point the client at it with `REMOTION_RENDER_API_URL=http://localhost:3000`.
 
+## Credits
+
+Every run charges credits. The CLI prints a footer on stdout when it does:
+
+```
+💳 Charged 31 credits · balance 1,240
+```
+
+Relay it to the user whenever it appears — it is the only signal they get about what a
+generation cost, and the balance is the only warning before a run fails with
+`insufficient_credits`. Do not drop it from your summary.
+
 ## Error handling
 
 - **`job_id` missing or not assets-ready**: the script exits non-zero with a message like "render job N is not in assets-ready state". The agent should call `prepare_video_assets` again with the updated DSL to regenerate, then re-call this skill with the new `job_id`.

package/skills/render-video/scripts/render_video.py

@@ -24,6 +24,7 @@
 """
 
 import argparse
+import atexit
 import builtins
 import json
 import os
@@ -356,6 +357,62 @@ def LogPrint(*args, sep=" ", end="\n", file=None, flush=False):
     builtins.print(f"[{stamp}] {message}", end=end, file=file, flush=flush)
 
 
+# ─── 扣费汇总 ────────────────────────────────────────────────────────────────
+# 一条视频管线的花费分散在多处：每个 gen-image / gen-voice / gen-video /
+# gen-digital-human 子进程各扣一次，渲染再通过 saveManifest 扣一次。子进程是
+# capture_output 起的，它们自己打的 💳 页脚会被吞掉，所以这里把子进程 stdout 里的
+# billing 事件抓出来累加，最后统一报一次总账——否则用户跑完一整条管线，只知道视频
+# 好了，不知道花了多少积分。
+_BILLING: dict = {"credits": 0, "balance": None}
+
+
+def absorb_child_billing(stdout: str) -> None:
+    """从子 skill 的 stdout 里收集 `__progress__` billing 事件（见 CLI src/billing.ts）。"""
+    for line in (stdout or "").splitlines():
+        line = line.strip()
+        if not line.startswith("{") or "__progress__" not in line:
+            continue
+        try:
+            event = json.loads(line)
+        except json.JSONDecodeError:
+            continue
+        if not isinstance(event, dict) or event.get("phase") != "billing":
+            continue
+        credits = event.get("credits")
+        if not isinstance(credits, (int, float)) or credits <= 0:
+            continue
+        _BILLING["credits"] += int(credits)
+        balance = event.get("balance")
+        if isinstance(balance, (int, float)):
+            _BILLING["balance"] = int(balance)
+
+
+def print_billing_footer() -> None:
+    """收尾时报一次总账。通过 atexit 注册，所以 sys.exit / 中途失败也会打印
+    ——已经发生的扣费不该因为后面某一步失败就不告知用户。
+
+    走 stdout 而不是 stderr：ab-agent 交给 LLM 的是 `result.stdout || result.stderr`
+    （mcp-tools.ts），只写 stderr 在托管 agent 里等于不可见。
+    """
+    credits = _BILLING["credits"]
+    balance = _BILLING["balance"]
+    try:
+        import render_job_client  # type: ignore
+
+        job_billing = render_job_client.billing_summary()
+        if job_billing.get("credits"):
+            credits += int(job_billing["credits"])
+            # saveManifest（渲染扣费）在管线里排最后，它带回的余额最新。
+            if job_billing.get("balance") is not None:
+                balance = int(job_billing["balance"])
+    except Exception:  # noqa: BLE001 — 报账失败绝不能影响已完成的渲染
+        pass
+    if credits <= 0:
+        return
+    suffix = f" · balance {balance:,}" if balance is not None else ""
+    builtins.print(f"\n💳 Charged {credits:,} credits{suffix}", flush=True)
+
+
 def sync_chrome_headless_vendor(renderer_dir: str, render_plan: dict) -> None:
     """Chrome Headless vendor 同步——实现已抽到独立模块 ``_chrome_vendor``。
 
@@ -758,6 +815,7 @@ def resolve_asset_image(asset: dict, private_token: str, timeout: int) -> dict:
 
     try:
         result = subprocess.run(cmd, capture_output=True, text=True, timeout=timeout, stdin=subprocess.DEVNULL)
+        absorb_child_billing(result.stdout)
         if result.returncode == 0:
             for line in reversed(result.stdout.strip().split("\n")):
                 line = line.strip()
@@ -809,6 +867,7 @@ def resolve_asset_audio(asset: dict, private_token: str, timeout: int) -> dict:
 
     try:
         result = subprocess.run(cmd, capture_output=True, text=True, timeout=timeout, stdin=subprocess.DEVNULL)
+        absorb_child_billing(result.stdout)
         if result.returncode == 0:
             for line in reversed(result.stdout.strip().split("\n")):
                 line = line.strip()
@@ -854,6 +913,7 @@ def resolve_asset_video(asset: dict, private_token: str, timeout: int) -> dict:
 
     try:
         result = subprocess.run(cmd, capture_output=True, text=True, timeout=timeout, stdin=subprocess.DEVNULL)
+        absorb_child_billing(result.stdout)
         if result.returncode == 0:
             for line in reversed(result.stdout.strip().split("\n")):
                 line = line.strip()
@@ -893,6 +953,7 @@ def resolve_asset_digital_human(asset: dict, private_token: str, timeout: int) -
     dh_timeout = max(timeout, 660)
     try:
         result = subprocess.run(cmd, capture_output=True, text=True, timeout=dh_timeout, stdin=subprocess.DEVNULL)
+        absorb_child_billing(result.stdout)
         if result.returncode == 0:
             for line in reversed(result.stdout.strip().split("\n")):
                 line = line.strip()
@@ -1892,6 +1953,8 @@ def _log_render_plan_summary(render_plan: dict) -> None:
 
 
 def main():
+    # 无论正常结束、sys.exit 还是中途失败，都在最后报一次积分账（无扣费时静默）。
+    atexit.register(print_billing_footer)
     parser = argparse.ArgumentParser(
         description="Video render tool — DSL + TemplateBinding → Remotion video.",
         formatter_class=argparse.RawDescriptionHelpFormatter,

package/skills/render-video/skill.json

@@ -5,6 +5,7 @@
   "category": "authoring",
   "title": "Remotion Video Renderer",
   "description": "Loads a persisted RenderPlan by job_id and drives the Remotion engine to produce the final video. Assets must already be generated via prepare_video_assets — this skill never resolves or regenerates assets.",
+  "auth": "required",
   "envVars": ["PRIV_TOKEN", "MM_API_BASE_URL", "MM_BACKEND_API_URL", "AGENT_NAME", "REMOTION_RENDER_API_URL", "REMOTION_RENDER_MODE", "REMOTION_OUTPUT_DIR", "ASSET_CACHE_DIR"],
   "scriptPath": "scripts/render_video.py",
   "parameters": {

package/skills/template-registry/scripts/list_templates.py

@@ -39,7 +39,13 @@ from pathlib import Path
 
 # When run as `python3 <skillDir>/scripts/list_templates.py`, this script's own
 # directory is sys.path[0], so the sibling shared modules import directly.
-from registry_loader import list_templates as load_visible_templates  # noqa: E402
+from registry_loader import (  # noqa: E402
+    EXIT_BACKEND_UNREACHABLE,
+    EXIT_NOT_AUTHENTICATED,
+    RegistryAuthError,
+    RegistryUnreachableError,
+    list_templates as load_visible_templates,
+)
 
 
 def _matches(tpl: dict, tag: str | None, aspect: str | None, language: str | None) -> bool:
@@ -166,6 +172,15 @@ def main() -> None:
 
     try:
         templates = load_visible_templates()
+    except RegistryAuthError as exc:
+        # 退出码 4 = 需要授权（见 registry_loader.EXIT_NOT_AUTHENTICATED），宿主
+        # 据此引导用户重新登录，而不是把它当成一次普通失败。
+        print(f"❌ {exc}", file=sys.stderr)
+        sys.exit(EXIT_NOT_AUTHENTICATED)
+    except RegistryUnreachableError as exc:
+        # 退出码 5 = 后端不可达；重试或修配置即可，不需要重新登录。
+        print(f"❌ {exc}", file=sys.stderr)
+        sys.exit(EXIT_BACKEND_UNREACHABLE)
     except RuntimeError as exc:
         print(f"❌ {exc}", file=sys.stderr)
         sys.exit(1)

package/skills/template-registry/scripts/registry_loader.py

@@ -19,7 +19,8 @@
        兼容 ab-agent 注入的 MM_BACKEND_API_URL。默认值内置在 CLI 自身，独立安装
        （codex / npm i -g）开箱即用，无需宿主在 spawn 时注入。
 
-  鉴权：PRIV_TOKEN（env）或 ~/.config/remixmate/credentials.json（同源条目）。
+  鉴权：只读 PRIV_TOKEN 环境变量。凭证库 / 钥匙串 / 设备授权流由 CLI 的 Node 侧
+  唯一鉴权入口负责（src/auth/ensure.ts），并在 spawn 时注入本进程环境。
 
   monorepo / 显式本地文件等其它来源已**移除**。需要离线 / pinned 数据的维护脚本
   （check_contracts.py / sync_registry.py）直接走 template_paths 读 monorepo，
@@ -47,7 +48,6 @@ import urllib.error
 import urllib.request
 from functools import lru_cache
 from typing import Optional
-from urllib.parse import urlparse
 
 __all__ = [
     "load_registry_data",
@@ -55,8 +55,29 @@ __all__ = [
     "get_template",
     "list_templates",
     "invalidate_cache",
+    "RegistryAuthError",
+    "RegistryUnreachableError",
+    "EXIT_NOT_AUTHENTICATED",
+    "EXIT_BACKEND_UNREACHABLE",
 ]
 
+# 与 Node 侧 src/errors.ts 的 EXIT 常量保持一致：宿主据此判断「该引导用户授权」
+# 还是「后端故障，等一会儿再试」，而不是把两者都当成一次普通失败。
+EXIT_NOT_AUTHENTICATED = 4
+EXIT_BACKEND_UNREACHABLE = 5
+
+
+class RegistryAuthError(RuntimeError):
+    """凭证缺失 / 失效 / 被吊销 —— 与「后端不可达」区分开的独立失败模式。
+
+    这两种情况以前被压成同一段「请确认 ab-api 可达…并已设置有效 PRIV_TOKEN」的
+    文案，用户无从判断该去登录还是该等后端恢复。
+    """
+
+
+class RegistryUnreachableError(RuntimeError):
+    """网络不通 / 超时 / 5xx —— 用户无需重新登录，重试或修配置即可。"""
+
 # 默认缓存 TTL（秒）。HTTP 拉取后该时间内不再重新请求。
 _DEFAULT_TTL_SECONDS = 300
 
@@ -127,58 +148,20 @@ def _write_cached(cache_file: str, payload: dict) -> None:
         raise
 
 
-# ── PrivToken 解析（env → 凭证库回退）──────────────────────────────────────
+# ── PrivToken 解析 ────────────────────────────────────────────────────────
 #
-# 与 CLI 的 resolve.ts 共享同一优先级与同一凭证文件路径（Req 7.5）。Python 侧只读
-# 文件存储（不读 keychain —— keychain 由 CLI 写/读，Python 用纯 stdlib 不引依赖）。
-
-def _credentials_file_path() -> str:
-    """凭证文件路径，与 CLI credential-store.ts 的 CRED_FILE 保持一致。"""
-    return os.path.join(os.path.expanduser("~"), ".config", "remixmate", "credentials.json")
-
-
-def _origin(url: str) -> tuple:
-    p = urlparse(url)
-    return (p.scheme, p.hostname, p.port)
-
+# 只读 PRIV_TOKEN 环境变量 —— 凭证库 / 系统钥匙串 / 设备授权流全部由 Node 侧的
+# 唯一鉴权入口（src/auth/ensure.ts）负责，并在 spawn 时把解析结果注入本进程的
+# 环境变量（src/runner.ts）。
+#
+# 这里曾经自己读 ~/.config/remixmate/credentials.json，带来两个问题：一是与 Node
+# 侧的解析逻辑各写一份、只有本 skill 享受得到；二是「只有一条凭证就直接用」的兜底
+# 会把某个后端（如本地开发环境）的 token 发往另一个后端（如生产），属于跨源泄露。
 
-def _priv_token_from_credential_store(url: str) -> str:
-    """从 ~/.config/remixmate/credentials.json 取与 url 同源的 PrivToken。
 
-    匹配顺序：① 与 registry URL 同源（scheme/host/port）的条目；② 仅有单条目时直接用。
-    secret=='keychain' 的条目跳过（Python 无法读取 keychain）。任何异常静默返回空串。
-    """
-    try:
-        with open(_credentials_file_path(), encoding="utf-8") as f:
-            doc = json.load(f)
-    except (OSError, json.JSONDecodeError):
-        return ""
-    creds = doc.get("credentials") if isinstance(doc, dict) else None
-    if not isinstance(creds, dict) or not creds:
-        return ""
-
-    target = _origin(url)
-    for base_url, entry in creds.items():
-        if not isinstance(entry, dict) or entry.get("secret") == "keychain":
-            continue
-        if _origin(base_url) == target:
-            tok = str(entry.get("privToken") or "").strip()
-            if tok:
-                return tok
-    # 单条目兜底：CLI 通常只登录一个后端。
-    if len(creds) == 1:
-        entry = next(iter(creds.values()))
-        if isinstance(entry, dict) and entry.get("secret") != "keychain":
-            return str(entry.get("privToken") or "").strip()
-    return ""
-
-
-def _resolve_priv_token(url: str) -> str:
-    """PrivToken 解析：环境变量优先，空则回退凭证库（与 CLI 一致的优先级）。"""
-    env = os.environ.get("PRIV_TOKEN", "").strip()
-    if env:
-        return env
-    return _priv_token_from_credential_store(url)
+def _resolve_priv_token(_url: str) -> str:
+    """PrivToken 解析：只认由 CLI 注入 / 宿主设置的 PRIV_TOKEN 环境变量。"""
+    return os.environ.get("PRIV_TOKEN", "").strip()
 
 
 def _fetch_http(url: str, timeout: int) -> dict:
@@ -214,9 +197,13 @@ def _fetch_http(url: str, timeout: int) -> dict:
 
             # ab-api 风格：{code, msg, data} → 取 data；其他场景视作直接 registry
             if isinstance(parsed, dict) and "data" in parsed and "code" in parsed:
-                if parsed.get("code") not in (0, "0"):
+                code = parsed.get("code")
+                if str(code) in ("401", "403"):
+                    # ab-api 用 HTTP 200 + code=401 表达鉴权失败，所以这里也要判。
+                    raise RegistryAuthError(str(parsed.get("msg") or f"code={code}"))
+                if code not in (0, "0"):
                     raise RuntimeError(
-                        f"registry HTTP 业务失败: code={parsed.get('code')} msg={parsed.get('msg')}"
+                        f"registry HTTP 业务失败: code={code} msg={parsed.get('msg')}"
                     )
                 data = parsed["data"]
             else:
@@ -233,6 +220,8 @@ def _fetch_http(url: str, timeout: int) -> dict:
             cached["fetchedAt"] = time.time()
             _write_cached(cache_file, cached)
             return cached["data"]
+        if e.code in (401, 403):
+            raise RegistryAuthError(f"HTTP {e.code}") from e
         raise
 
 
@@ -249,6 +238,13 @@ def _load_from_url(url: str, ttl: int, timeout: int) -> dict:
 
     try:
         return _fetch_http(url, timeout)
+    except RegistryAuthError as exc:
+        # 鉴权失败不降级到缓存：缓存可能属于另一个账号，静默返回它会把「凭证已失效」
+        # 掩盖成一份看似正常、实则不属于当前用户的模板列表。
+        raise RegistryAuthError(
+            f"凭证无效或已过期（{exc}）。运行 remixmate login 重新授权，"
+            f"或设置有效的 PRIV_TOKEN。当前后端：{url}"
+        ) from exc
     except Exception as exc:
         # HTTP 故障时降级：有磁盘缓存就继续用（同一数据源的容错，不是第二个源）
         if cached:
@@ -257,10 +253,9 @@ def _load_from_url(url: str, ttl: int, timeout: int) -> dict:
                 file=sys.stderr,
             )
             return cached["data"]
-        raise RuntimeError(
-            f"registry HTTP 拉取失败且无可用缓存: {url} ({exc})。"
-            f" 请确认 ab-api 可达、VIDEO_TEMPLATE_REGISTRY_URL / MM_API_BASE_URL"
-            f" 配置正确，并已设置有效 PRIV_TOKEN。"
+        raise RegistryUnreachableError(
+            f"registry 后端不可达且无可用缓存: {url} ({exc})。"
+            f" 请确认 ab-api 可达、VIDEO_TEMPLATE_REGISTRY_URL / MM_API_BASE_URL 配置正确。"
         ) from exc
 
 
@@ -275,16 +270,21 @@ def load_registry_data(
     未设置则按 ``MM_API_BASE_URL`` 推导（见 ``_default_registry_url``）。
     带磁盘缓存（TTL + ETag/304），返回 ``data.get("templates", [])`` 形状的 dict。
     """
+    # 这两行是实现细节（endpoint 从哪来），对终端用户没有意义，却出现在每一次
+    # 成功调用的 stderr 上。默认收起，排障时用 REMIXMATE_DEBUG=1 打开。
+    debug = os.environ.get("REMIXMATE_DEBUG", "").strip() == "1"
     url = os.environ.get("VIDEO_TEMPLATE_REGISTRY_URL", "").strip()
     if url:
-        print(f"[registry_loader] using HTTP registry: {url}", file=sys.stderr)
+        if debug:
+            print(f"[registry_loader] using HTTP registry: {url}", file=sys.stderr)
     else:
         url = _default_registry_url()
-        print(
-            f"[registry_loader] VIDEO_TEMPLATE_REGISTRY_URL unset; using derived"
-            f" default URL: {url}",
-            file=sys.stderr,
-        )
+        if debug:
+            print(
+                f"[registry_loader] VIDEO_TEMPLATE_REGISTRY_URL unset; using derived"
+                f" default URL: {url}",
+                file=sys.stderr,
+            )
     return _load_from_url(url, ttl=ttl_seconds, timeout=http_timeout)
 
 

package/skills/template-registry/scripts/render_job_client.py

@@ -37,6 +37,31 @@ def _make_headers(priv_token: str) -> dict:
     return headers
 
 
+# ─── 扣费回传 ────────────────────────────────────────────────────────────────
+# ab-api 会在扣了积分的成功响应上挂 billing 字段（见 core.Success）。渲染积分是在
+# /renderJob/saveManifest 里扣的，所以这里是渲染花费唯一的可见入口——累加起来交给
+# render_video.py 在收尾时统一告知用户。
+_BILLING: dict = {"credits": 0, "balance": None}
+
+
+def _record_billing(result: dict) -> None:
+    billing = result.get("billing")
+    if not isinstance(billing, dict):
+        return
+    credits = billing.get("credits")
+    if not isinstance(credits, (int, float)) or credits <= 0:
+        return
+    _BILLING["credits"] += int(credits)
+    balance = billing.get("balance")
+    if isinstance(balance, (int, float)):
+        _BILLING["balance"] = int(balance)
+
+
+def billing_summary() -> dict:
+    """本进程通过本模块产生的累计扣费：{"credits": int, "balance": int|None}。"""
+    return dict(_BILLING)
+
+
 def _post(path: str, payload: dict, priv_token: str, timeout: int = 30) -> dict:
     """发送 POST 请求并返回响应 data 字段；失败时抛出 RuntimeError。"""
     url = f"{_api_base()}{path}"
@@ -50,6 +75,8 @@ def _post(path: str, payload: dict, priv_token: str, timeout: int = 30) -> dict:
     except urllib.error.URLError as e:
         raise RuntimeError(f"network request failed ({path}): {e.reason}") from e
 
+    _record_billing(result)
+
     if result.get("code") != 0:
         msg = result.get("msg") or result.get("message") or "unknown error"
         raise RuntimeError(f"API returned an error [{path}]: {msg}")

package/skills/template-registry/skill.json

@@ -5,6 +5,7 @@
   "category": "authoring",
   "title": "Video Template Registry",
   "description": "List all available video templates (templateId / name / aspect ratio / style tags). Template-to-DSL binding is no longer exposed as a separate step — once prepare_video_assets receives a template_id it runs the full DSL→RenderPlan pipeline internally.",
+  "auth": "required",
   "envVars": ["PRIV_TOKEN", "VIDEO_TEMPLATE_REGISTRY_URL", "MM_API_BASE_URL", "VIDEO_TEMPLATE_REGISTRY_HTTP_METHOD"],
   "entry": { "type": "python", "scriptPath": "scripts/list_templates.py" },
   "parameters": {

package/skills/video-parser/skill.json

@@ -5,6 +5,7 @@
   "category": "consuming",
   "title": "Video Deconstruction",
   "description": "Video deconstruction: split a video into reusable assets (audio extraction, ASR transcript, keyframe extraction, scene segmentation).",
+  "auth": "required",
   "envVars": ["RENDER_API_URL", "PRIV_TOKEN", "CONVERSATION_ID"],
   "scriptPath": "scripts/parse_via_render.py",
   "parameters": {

package/skills/web-record/skill.json

@@ -5,6 +5,7 @@
   "category": "consuming",
   "title": "Web Page Recording",
   "description": "Drive a headless browser (Playwright Python) to RECORD any URL to a video, then (by default) transcode to mp4, grab a cover frame, upload to VOD and return a playable CDN URL. Modes: fixed-duration recording, condition-triggered stop (element appears / disappears), auto-scroll from top to bottom, custom storyboards, and parameterized templates. Storyboard scenes: highlight / focus / zoom / scroll / virtual-cursor click / type / hover / caption / title-card / arrow / numbered sequence / redact / code-line highlight. For still images (png/jpg) use the web_screenshot tool instead. Use this tool whenever the user wants a video / recording / screencast of a web page: record a page, scroll-through video, page-from-top-to-bottom clip, demo of clicks/typing/hover, storyboard / multi-scene intro video, or a template-based clip.",
+  "auth": "optional",
   "envVars": ["WEB_CAPTURE_BROWSER", "WEB_CAPTURE_OUTPUT_BASE", "WEB_CAPTURE_NO_SANDBOX", "WEB_CAPTURE_ALLOW_PRIVATE_HOSTS", "PLAYWRIGHT_BROWSERS_PATH", "REMOTION_RENDER_API_URL", "PRIV_TOKEN"],
   "entry": { "type": "python", "scriptPath": "../web-screenshot/scripts/record.py" },
   "parameters": {

package/skills/web-screenshot/skill.json

@@ -5,6 +5,7 @@
   "category": "consuming",
   "title": "Web Page Screenshot",
   "description": "Drive a headless browser (Playwright Python) to capture any URL to a local STILL IMAGE (png/jpg): full-page / viewport / element / region screenshots, with device emulation, waiting, hide/mask/redact, and static annotations (highlight / arrow / caption / numbered sequence / redact). Images only — for video / recording / scroll-through / storyboard clips (webm) use the web_record tool instead.",
+  "auth": "none",
   "envVars": ["WEB_CAPTURE_BROWSER", "WEB_CAPTURE_OUTPUT_BASE", "WEB_CAPTURE_NO_SANDBOX", "WEB_CAPTURE_ALLOW_PRIVATE_HOSTS", "PLAYWRIGHT_BROWSERS_PATH"],
   "entry": { "type": "python", "scriptPath": "scripts/screenshot.py" },
   "parameters": {