feat(v1.8.1): Windows 兼容性修复 + SKILL.md 铁律区块

Windows 兼容性修复（基于真实使用痛点）:

- force_utf8_stdout(): 强制 stdout/stderr 为 UTF-8，修复 Windows GBK 崩溃（print('\\xa0') 不再炸）

- resolve_instances/load_config 新增 %APPDATA%/searxng-cli/ 路径，覆盖 Windows 配置约定

- fetch.py 失败诊断增强：输出 status_code=/cause=/url= 字段，AI Agent 可程序化区分 404/403/DNS 失败

SKILL.md 铁律区块（5 条，置顶）:

- stdout=数据/stderr=日志 永不混淆

- 禁用 2>/dev/null（丢弃 stderr = 失败时零诊断）

- 排错去 --quiet 加 --verbose

- 配置查找覆盖 WSL + Windows 双路径

- 实例 URL 必填，公共实例发现已移除

测试: 352 -> 362（新增 10 个：force_utf8_stdout 幂等性/GBK 替换/非 ASCII 打印/APPDATA 路径发现/txt 回退/空 APPDATA）

SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: searxng-use-cli
 description: Use when you need to search the web via your OWN SearXNG instance (no public-instance discovery). 3 CLI scripts + a shared common.py module — execute privacy-respecting searches against a user-supplied instance (with multi-instance failover, 5xx/429 retry, auto-fetch) or via SEARXNG_INSTANCE env / config file, fetch/extract readable text or markdown from web pages. Zero-config replacement for proprietary search APIs.
-version: 1.8.0
+version: 1.8.1
 author: Metona Team
 license: MIT
 platforms: [linux, macos, windows]
@@ -13,6 +13,20 @@ metadata:
 
 # SearXNG CLI Toolkit
 
+## 铁律（AI Agent 必读，违反会导致致命后果）
+
+> **这些规则优先级最高，必须在任何调用前遵守。**
+
+1. **stdout=数据，stderr=日志，永不混淆。** stdout 只输出 JSON/CSV/文本数据，stderr 只输出日志/进度/警告。AI 解析 stdout，人类看 stderr。违反：AI 会把日志当数据解析，结果错乱。
+
+2. **绝对不要 `2>/dev/null` 或重定向 stderr 到 stdout。** stderr 携带排错关键信息（重试日志、HTTP 状态码、缓存命中、认证警告）。丢弃 stderr = 失败时零诊断信息，无法定位原因。需要静默时用 `--quiet`（仅抑制进度，保留 WARNING+ERROR），不要丢弃 stderr。
+
+3. **排错时第一步：去掉 `--quiet`，加 `--verbose`。** `--quiet` 只留 WARNING 级别，会吞掉 INFO 级别的重试日志、缓存命中提示、实例切换记录。诊断失败时必须用 `--verbose` 看到完整 HTTP 请求/响应/重试链。
+
+4. **配置查找必须覆盖双路径（WSL + Windows）。** Windows 环境下配置可能在 `~/.config/searxng-cli/`（WSL HOME）或 `%APPDATA%/searxng-cli/`（Windows APPDATA）。检查配置存在性时两个路径都要查，否则会误判"无配置"并反问用户。
+
+5. **实例 URL 必填，公共实例发现已移除。** 必须通过 `-i`、`SEARXNG_INSTANCE` 环境变量、或配置文件提供实例 URL。无实例时 `search.py` 报 `E_CONFIG` 退出，不要尝试猜测或硬编码公共实例。
+
 ## Overview
 
 SearXNG is a privacy-respecting metasearch engine that aggregates results from 70+ search services without tracking users. This skill provides three standalone Python CLI scripts — works with **any AI agent** (Hermes, Claude Code, Codex, OpenCode, Cursor, Trae, etc.) or directly from your terminal.
@@ -40,7 +54,7 @@ SearXNG is a privacy-respecting metasearch engine that aggregates results from 7
 **Caching & config**
 - SQLite result caching (`--cache-ttl`) — identical queries within a TTL skip the network entirely; `--clear-cache` / `--cache-stats` manage it
 - Config file (`searxng.toml`) pre-sets most flags; `--config FILE` loads a non-default config; `instances.txt` for plain URL lists
-- Instance resolution priority: `-i` → `SEARXNG_INSTANCE` env → config file
+- Instance resolution priority: `-i` → `SEARXNG_INSTANCE` env → config file (`./searxng.toml` → `~/.config/searxng-cli/searxng.toml` → `%APPDATA%/searxng-cli/searxng.toml` on Windows)
 
 **Network & auth**
 - Proxy support (`--proxy`) for both search and fetch (sets `HTTP_PROXY`/`HTTPS_PROXY`/`NO_PROXY`)
@@ -52,6 +66,9 @@ SearXNG is a privacy-respecting metasearch engine that aggregates results from 7
 - Shared `common.py` module — unified retry/charset/auth/logging logic across both scripts
 - `search.py --fetch` reuses `fetch.py`'s higher-quality text extractor (no code duplication)
 - Structured logging (`--verbose` / `--quiet`) — three levels: default INFO (progress + warnings), `--verbose` DEBUG (HTTP detail, cache keys), `--quiet` WARNING (errors only). All log output to stderr; stdout reserved for data
+- UTF-8 stdout enforcement (`force_utf8_stdout()`) — Windows Python defaults to GBK and crashes on non-ASCII chars; both scripts force UTF-8 + `errors='replace'` at startup so `print('\xa0')` never raises
+- Windows config discovery — `resolve_instances` / `load_config` also check `%APPDATA%/searxng-cli/` (Windows per-user app convention) in addition to `~/.config/searxng-cli/` (POSIX convention)
+- fetch.py failure diagnostics — error output includes `status_code=`, `cause=`, `url=` fields so AI agents can programmatically distinguish 404 vs 403 vs DNS failure without parsing English prose
 - Engine/category whitespace normalization (`"google, bing"` → `"google,bing"`)
 - `--time-range none` option to disable time filtering
 
@@ -98,7 +115,8 @@ python scripts/fetch.py -u "https://example.com" --extract text
 export SEARXNG_INSTANCE="https://my-searxng.example.com,https://backup.example.com"
 python scripts/search.py -q "python asyncio tutorial"          # -i not needed
 
-# 6. Or use a config file (./searxng.toml or ~/.config/searxng-cli/searxng.toml)
+# 6. Or use a config file (./searxng.toml or ~/.config/searxng-cli/searxng.toml
+#    or %APPDATA%/searxng-cli/searxng.toml on Windows)
 #    [searxng]
 #    instance = "https://my-searxng.example.com"
 #    # or: instances = ["https://a.example.com", "https://b.example.com"]
@@ -403,7 +421,7 @@ usage: search.py [-h] [--query QUERY] [--instance URL]
 - `--max-results N` — limit number of results (applied AFTER dedup+sort, so the highest-scoring/newest items are kept)
 - `--sort-by {score,date,engine,none}` — sort results (default: `score` descending; `none` preserves instance order). Applied after dedup, before `--max-results`. HTML-fallback results have no score and keep their order
 - `--no-dedup` — disable cross-engine deduplication (by default, duplicate URLs — same page ignoring tracking params/fragment — are collapsed, keeping the first occurrence's engine/score)
-- `--config FILE` — path to a `searxng.toml` config file; overrides the default auto-discovery (`./searxng.toml` → `~/.config/searxng-cli/searxng.toml`). Must be the first flag so its values can set defaults for other flags
+- `--config FILE` — path to a `searxng.toml` config file; overrides the default auto-discovery (`./searxng.toml` → `~/.config/searxng-cli/searxng.toml` → `%APPDATA%/searxng-cli/searxng.toml` on Windows). Must be the first flag so its values can set defaults for other flags
 - `--verbose` / `-v` — show debug-level diagnostics on stderr (HTTP request URLs, response codes, cache keys, retry detail)
 - `--quiet` — suppress progress messages and retry notices on stderr; only warnings and errors are shown (no short flag: `-q` is `--query`)
 - `--include-domain a.com,b.org` — allowlist; only results from these domains are kept (applied after search)