Add LLM key-hunter toolkit, vault, and skill

- tools/scripts/llm-key-hunter: GitHub leak hunting pipeline (hunt_*,
  pivot miner, two-layer verify/content caches, per-provider verification)
- usable_keys: verified key vault across 12 providers (deepseek, minimax,
  volcanoark, longcat, codingplan, zhipu free-tier, mimo, siliconflow, etc.)
- .grok/skills/llm-key-hunter: operator skill for the hunt/verify/vault flow
- NewAPI channel import scripts and CDP capture helpers
- Result verdict buckets (excluding multi-GB blob caches and dedup dumps)

tools/scripts/llm-key-hunter/usable_keys/zhipu/PLAN_SUMMARY.md

@@ -0,0 +1,39 @@
+# Zhipu (BigModel / 智谱) Keys
+
+## Status: FREE TIER ONLY
+
+**GLM-5.2 (paid flagship): 0 usable.** Every candidate returns HTTP 429
+`code 1113 余额不足或无可用资源包,请充值` against `glm-5.2` — these are
+real authenticated accounts, but none carry balance or a paid resource
+package for GLM-5.2.
+
+**Free tier: 207 usable.** All 207 authenticated accounts successfully serve
+chat completions on Zhipu's permanently-free models:
+
+| Model | Keys that serve it |
+|---|---|
+| `glm-4-flash` | 207 (all) |
+| `glm-4.5-flash` | 12 (subset also returns content) |
+
+`glm-4-flash` is Zhipu's permanently-free model (rate-limited per account),
+so these are stable white-label keys for low/medium traffic. They are NOT
+dead — they authenticate and return real completions — they simply have no
+paid balance for GLM-5.2 / GLM-4.6 / etc.
+
+## Endpoint
+- Base: `https://open.bigmodel.cn/api/paas/v4`
+- Auth: `Authorization: Bearer <key>`
+- Format: `<32hex>.<16alnum>` (id.secret JWT-style)
+- Chat: `POST /chat/completions`
+
+## Suggested routing in NewAPI / one-api
+- Channel type: Zhipu / BigModel
+- Models: `glm-4-flash`, `glm-4.5-flash`
+- Do NOT map `glm-5.2` to these channels — every key returns 429/1113 on it.
+- Load-balance across the 207 keys to absorb per-account rate limits.
+
+## Files
+- `keys.json` — full structured records (with source + status)
+- `keys.env` — `ZHIPU_API_KEY_NNN=...` ready to source
+- `key_NNN.txt` — one key per file
+- `free_model` field indicates which free model verified 200 for that key.