feat(webfetch): add reusable ten-method native configuration

.env.example

@@ -6,6 +6,8 @@ OPENCLAW_REPO=
 OPENCLAW_NODE=~/.browser-eval/toolchains/openclaw-node/node_modules/node/bin/node
 OPENCLAW_PNPM=~/.browser-eval/toolchains/openclaw-node/node_modules/.bin/pnpm
 DEEPSEEK_HARNESS_REPO=
+# DeepSeek API (Fetch and Search): https://platform.deepseek.com/api_keys
+DEEPSEEK_API_KEY=
 
 OSS_PYTHON=.venv/bin/python
 # Official Moli CLIs; each measured release binary is version- and SHA-256-gated.
@@ -72,7 +74,9 @@ FIRECRAWL_API_KEY=
 TAVILY_API_KEY=
 EXA_API_KEY=
 BROWSERBASE_API_KEY=
+# Optional for browser sessions; not required by the Fetch adapter.
 BROWSERBASE_PROJECT_ID=
+BROWSER_USE_API_KEY=
 # Independent SERP comparison: https://serper.dev/ and https://brightdata.com/
 SERPER_API_KEY=
 BRIGHTDATA_API_KEY=
@@ -97,9 +101,10 @@ OXYLABS_PASSWORD=
 # Create it in Rainbond: Personal Center -> Access Token.
 RAINBOND_API_TOKEN=
 
+# Crawl4AI Cloud: https://crawl4ai.com/dashboard/ (distinct from the local library)
+CRAWL4AI_CLOUD_API_KEY=
 # Web Search providers. Keep evaluation credentials in this repository file.
 LEXMOUNT_SEARCH_API_KEY=
-DEEPSEEK_API_KEY=
 BOCHA_API_KEY=
 # Optional, only for explicitly selected model relevance evaluation.
 SEARCH_JUDGE_API_KEY=

README.md

@@ -45,6 +45,8 @@ Open the report path printed in the terminal. The example has one expected pass
 
 Next, [try HTTP, Moli or Chromium on a local article](examples/web_fetch/README.md). Optional adapters, service credentials and live evaluation commands are covered in the track guides above.
 
+The reusable ten-method Web Fetch cohort is selected with `--providers configured --provider-config configs/providers.example.json`; shared defaults are 30 seconds per call and one failure retry. See the [provider contract](docs/provider-contract.md#selected-native-cohort).
+
 ## Reports
 
 Reports include comparisons, task results, full responses and failure evidence. Runs are saved under `~/.browser-eval/runs/`; use `--output /absolute/path/new-run` to choose another directory.

configs/providers.example.json

@@ -1,6 +1,61 @@
 {
   "schema_version": 1,
+  "default_providers": [
+    "moli-webfetch",
+    "firecrawl",
+    "cloudflare-markdown",
+    "tavily",
+    "exa",
+    "brightdata",
+    "browser-use-fetch",
+    "browserbase",
+    "deepseek-fetch-api",
+    "crawl4ai-cloud"
+  ],
   "providers": {
+    "crawl4ai-cloud": {
+      "type": "crawl4ai_cloud",
+      "api_key": "${CRAWL4AI_CLOUD_API_KEY}",
+      "engine": "browser",
+      "proxy": "residential",
+      "output_format": "markdown",
+      "dns_mode": "system",
+      "log_level": "debug",
+      "uniform_native_parameters": true
+    },
+    "deepseek-fetch-api": {
+      "type": "deepseek_fetch",
+      "api_key": "${DEEPSEEK_API_KEY}",
+      "dns_mode": "system",
+      "log_level": "debug",
+      "uniform_native_parameters": true
+    },
+    "brightdata": {
+      "type": "brightdata_unlocker",
+      "api_key": "${BRIGHTDATA_API_TOKEN}",
+      "zone": "${BRIGHTDATA_UNLOCKER_ZONE}",
+      "render": true,
+      "debug": true,
+      "log_level": "debug",
+      "dns_mode": "system",
+      "uniform_native_parameters": true
+    },
+    "browser-use-fetch": {
+      "type": "browser_use_fetch",
+      "api_key": "${BROWSER_USE_API_KEY}",
+      "proxy_country": "US",
+      "log_level": "debug",
+      "dns_mode": "system",
+      "uniform_native_parameters": true
+    },
+    "browserbase": {
+      "type": "browserbase_fetch",
+      "api_key": "${BROWSERBASE_API_KEY}",
+      "proxies": true,
+      "log_level": "debug",
+      "dns_mode": "system",
+      "uniform_native_parameters": true
+    },
     "moli-webfetch": {
       "type": "moli_webfetch",
       "binary": "${MOLI_BINARY:-moli}",
@@ -10,14 +65,11 @@
       "expected_binary_sha256": "7a53cdd071ae60a2f8ff6d08107f67f6b0f1d6247a9acd970d2bcec77ff15b50",
       "wait_until": "done",
       "block_private_networks": true,
-      "private_network_allowlist_hosts": [
-        "browser.lexmount.com",
-        "react-shopping-cart-67954.firebaseapp.com"
-      ],
       "max_bytes": 20000000,
       "log_level": "debug",
       "retain_complete_diagnostic_log": true,
-      "max_diagnostic_bytes": 100000000
+      "max_diagnostic_bytes": 100000000,
+      "uniform_native_parameters": true
     },
     "fixture": {
       "type": "fixture",
@@ -53,14 +105,15 @@
     },
     "firecrawl": {
       "type": "firecrawl",
+      "log_level": "debug",
       "base_url": "https://api.firecrawl.dev",
       "api_key": "${FIRECRAWL_API_KEY:-}",
       "max_age": 0,
       "max_bytes": 20000000,
       "min_interval_seconds": 6.1,
       "proxy": "auto",
       "capacity_credits_per_call": 1,
-      "dynamic_wait_strategy": "actions"
+      "uniform_native_parameters": true
     },
     "firecrawl-resources": {
       "type": "firecrawl",
@@ -76,11 +129,12 @@
     },
     "tavily": {
       "type": "tavily_extract",
+      "log_level": "debug",
       "base_url": "https://api.tavily.com",
       "api_key": "${TAVILY_API_KEY}",
       "extract_depth": "advanced",
-      "extract_timeout_seconds": 60.0,
-      "max_bytes": 20000000
+      "max_bytes": 20000000,
+      "uniform_native_parameters": true
     },
     "tavily-resources": {
       "type": "tavily_extract",
@@ -93,12 +147,14 @@
     },
     "exa": {
       "type": "exa_contents",
+      "log_level": "debug",
       "base_url": "https://api.exa.ai",
       "api_key": "${EXA_API_KEY}",
       "text_verbosity": "full",
       "max_age_hours": 0,
-      "livecrawl_timeout_ms": 15000,
-      "max_bytes": 20000000
+      "max_bytes": 20000000,
+      "uniform_native_parameters": true,
+      "rich_links_limit": 100
     },
     "exa-resources": {
       "type": "exa_contents",
@@ -112,11 +168,13 @@
     },
     "cloudflare-markdown": {
       "type": "cloudflare_markdown",
+      "log_level": "debug",
       "api_token": "${CLOUDFLARE_API_TOKEN}",
       "account_id": "${CLOUDFLARE_ACCOUNT_ID}",
       "max_bytes": 20000000,
       "document_max_bytes": 20000000,
-      "wait_until": "networkidle2"
+      "wait_until": "networkidle2",
+      "uniform_native_parameters": true
     },
     "cloudflare-content": {
       "type": "cloudflare_content",

docs/README.zh-CN.md

@@ -45,6 +45,8 @@ uv run --frozen browser-harness-eval run --delay 0
 
 接下来可以[用 HTTP、Moli 或 Chromium 读取本地文章](../examples/web_fetch/README.md)。可选适配器、服务凭据和在线评测命令见上面的各流程指南。
 
+使用 `--providers configured --provider-config configs/providers.example.json` 选择可复用的十种 Web Fetch 方法；默认每次调用 30 秒，失败允许重试一次。详见[适配器规则](provider-contract.md#selected-native-cohort)。
+
 ## 报告
 
 报告包含结果对比、逐题结果、完整响应和失败证据。运行结果默认保存在 `~/.browser-eval/runs/`，也可用 `--output /absolute/path/new-run` 指定新目录。

docs/provider-contract.md

@@ -14,6 +14,13 @@ fetch(case: EvalCase, timeout: float) -> ProviderResult
 - provider-specific request ID, engine, server elapsed time and billing data;
 - a stable error type and actionable error message.
 
+Disclose service API status as `metadata.api_http_status` and the final target-page
+status as `metadata.upstream_status`; use `None` when unavailable. Do not copy an
+API status into the target status. Before scoring and retry selection, the shared
+answer boundary marks a known non-2xx target response as a failed call, preserving
+its returned content and existing failure details. Content acceptance and call
+success remain separate requirements; historical results are not rewritten.
+
 Credentials are expanded from environment placeholders in the provider config.
 They must never be added to `metadata`, content, manifest or logs.
 
@@ -25,6 +32,42 @@ They must never be added to `metadata`, content, manifest or logs.
 
 Use the existing scorer and report renderer.
 
+### Hosted Markdown fetch services
+
+Five methods share `providers/hosted_fetch.py` and the existing HTTP client,
+runner and scorer. Their credentials and fixed options are declared in
+`configs/providers.example.json`; credentials come from the repository `.env`.
+
+| Method | Credentials | Native route and example options |
+| --- | --- | --- |
+| Bright Data | `BRIGHTDATA_API_TOKEN`, `BRIGHTDATA_UNLOCKER_ZONE` | `/request`, raw Markdown, rendering and native debug enabled |
+| Browser Use Fetch | `BROWSER_USE_API_KEY` | `/fetch`, Markdown, fresh session, US proxy; no Agent |
+| Browserbase Fetch | `BROWSERBASE_API_KEY` | `/v1/fetch`, Markdown, redirects and native proxy enabled; no Project ID |
+| DeepSeek Fetch API | `DEEPSEEK_API_KEY` | `/v1/tools/fetch`, URL-only request; retain content and title |
+| Crawl4AI Cloud | `CRAWL4AI_CLOUD_API_KEY` | `/scrape`, `format=md`, `engine=browser`, `proxy=residential` |
+
+These adapters implement webpage body/structure tasks. Unsupported document,
+media and file tasks fail before transport, without helper downloads or a
+substitute browser/Agent. Rendering and internal retries remain service-owned;
+offline adapter checks do not establish JavaScript capability or task pass rates.
+
+Retain complete native responses separately from score-bearing answer fields.
+An API HTTP 200 does not prove target retrieval succeeded: explicit service or
+origin errors, empty content and malformed responses remain failures. API status
+is distinct from target-site status. `log_level: debug` saves redacted caller
+requests, response headers/body and errors; it cannot expose undisclosed service
+internals. Only Bright Data receives the native debug option.
+
+`dns_mode: system` uses the local resolver for API hosts; target-site DNS belongs
+to the service. No automatic DNS fallback is added. All five services retain their
+native internal timeout defaults within the caller's deadline. Historical runs
+keep their recorded configuration.
+
+API references: [Bright Data](https://docs.brightdata.com/api-reference/rest-api/unlocker/unlock-website),
+[fetch-use](https://browser-use.com/posts/fetch-use),
+[Browserbase Fetch](https://docs.browserbase.com/platform/fetch/overview),
+[Crawl4AI Cloud](https://api.crawl4ai.com/docs/).
+
 | Output contract | Comparison scope |
 | --- | --- |
 | Cleaned content: `main_text`, Markdown or text | Shared readable-content leaderboard |
@@ -91,3 +134,45 @@ The version-2 extraction scope describes text extraction and requires the
 dynamic text route. Combining it with resource discovery is rejected before a
 call. Direct file delivery also rejects DOM completion: it cannot claim a
 rendered-page condition was satisfied while fetching only response bytes.
+
+## Selected native cohort
+
+`configs/providers.example.json` owns the ten retained methods in
+`default_providers`. Use `--providers configured --provider-config
+configs/providers.example.json` with `run`, `validate`, `run_benchmark.py` or
+`run_certified_benchmark.py`. Selection is frozen as explicit names in evidence;
+other adapters remain selectable, and the no-argument CLI still uses a fixture.
+
+- The default is 30 seconds per physical call and one failure retry. The final
+  response supplies the result and primary latency; every call and accumulated
+  retry cost is retained. CLI overrides remain explicit and recorded.
+- Stop timing after the response and required decoding; preparation for scoring,
+  scoring and evidence persistence follow. Discard late answers and quarantine
+  still-running workers rather than overlapping abandoned calls.
+- Reject per-case hints and never derive requests from grading requirements.
+  Dynamic cases declare native retrieval mode. Exa uniformly requests 100 rich
+  links; Cloudflare receives URLs directly, without helper document downloads.
+- Keep the 20,000,000-byte HTTP response/Moli output ceiling and debug logs.
+  Moli retains its pinned binary identity and private-network guard without
+  per-host exceptions or a caller-imposed subresource download cap. Non-public
+  DNS answers must be resolved as an environment issue before a public cohort.
+- `crawl4ai-cloud` is distinct from local `crawl4ai`: browser versions and
+  extraction differ, and old local results must not be relabeled as Cloud.
+  Cloud uses service-cleaned Markdown; retain its engine/cache/usage fields.
+  Its configured native proxy adds provider charges; never switch proxy policy
+  in response to scores.
+
+For diagnostics, supply frozen tasks and a new external output. `--measure-only`
+retains scored evidence without building a report; it does not certify a cohort.
+Historical unscored collections remain ineligible for scored reports.
+
+```sh
+uv run --frozen python scripts/run_benchmark.py \
+  --dataset /path/to/frozen-dataset.json \
+  --provider-config configs/providers.example.json --providers configured \
+  --timeout 30 --max-failure-retries 1 --workers 1 --measure-only \
+  --output /path/to/new-run
+```
+
+Publishable cohorts use `scripts/run_certified_benchmark.py` with its required
+scope, work root and output. Report profiles and measured data stay outside Git.

harness/record_io.py

@@ -4,6 +4,11 @@
 from pathlib import Path
 
 
+def parse_utf8_json(body: bytes):
+    """Use the same strict wire encoding for acquisition and evidence replay."""
+    return json.loads(body.decode('utf-8'))
+
+
 def file_sha256(path):
     """Hash file bytes without loading the entire artifact into memory.
 

harness/web_fetch/artifacts.py

@@ -164,9 +164,12 @@ def verify_public_call_log(root: Path, binding: dict[str, Any], *, allow_public_
                    for event in events)
             or events[0]["stage"] != "call_started"
             or events[-1]["stage"] != "call_completed"
-            or not any(event["stage"] in {"response_received", "exception_raised"}
+            or not any(event["stage"] in {"response_received", "exception_raised", "answer_deadline_exceeded"}
                        for event in events[1:-1])):
         raise ValueError("Public call log has an unexpected event schema")
+    if any(event["stage"] == "answer_deadline_exceeded" and event["error_type"] != "timeout"
+           for event in events):
+        raise ValueError("Deadline event must identify a timeout")
     summary = payload["request"]["summary"]
     if summary is not None and not isinstance(summary, dict):
         raise ValueError("Public call log has an unexpected request summary")

harness/web_fetch/benchmark_runner.py

@@ -15,6 +15,7 @@
     load_dataset,
     load_provider_config,
     load_runtime_env,
+    resolve_provider_selection,
 )
 from harness.web_fetch.network_preflight import preflight_guarded_providers
 from harness.web_fetch.providers import build_provider
@@ -33,7 +34,7 @@ def main(argv=None):
     parser.add_argument('--env-file',type=Path)
     parser.add_argument('--provider-config',type=Path,
                         help='One provider configuration for all selected tasks; defaults to the repository task-group configurations')
-    parser.add_argument('--providers',help='Comma-separated names; default: all benchmark methods')
+    parser.add_argument('--providers',help='Comma-separated names or configured; default: dataset coverage plan')
     parser.add_argument('--track',help='Run only this task group, such as dynamic; default: all groups')
     parser.add_argument('--case-ids',help='Comma-separated case IDs; default: every case in the selected task group')
     parser.add_argument('--validate-only',action='store_true',help='Check selected configuration without creating files or calling providers')
@@ -56,6 +57,10 @@ def main(argv=None):
     if args.runs < 1:parser.error('--runs must be positive')
     if not args.validate_only and args.output.exists():parser.error('Output exists; choose an empty directory to avoid overwriting measured evidence')
     load_runtime_env(args.env_file)
+    try:
+        args.providers = resolve_provider_selection(args.providers, args.provider_config)
+    except ValueError as error:
+        parser.error(str(error))
     dataset,cases=load_dataset(args.dataset)
     if args.track:
         cases=[c for c in cases if c.track==args.track]