lidge-jun/codexclaw / plugins/codexclaw/skills/search
cxc-search
VerifiedUse for web/current lookups. Triggers: search, look up, latest, news, prices, docs, status, X/Twitter, deep research, 검색, 검색해, 찾아봐, 찾아줘, 알아봐, 웹검색, 딥리서치, 심층 조사.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 3.7 KB |
| agents/openai.yaml | config | 157 B |
| references/blocked-url-reader.md | markdown | 1.9 KB |
| references/deep-research-entry.md | markdown | 1.7 KB |
| references/deep-research.md | markdown | 9.0 KB |
| references/discovery-boundaries.md | markdown | 887 B |
| references/divergence-grounding.md | markdown | 1.3 KB |
| references/hosted-discovery.md | markdown | 559 B |
| references/intent-guard.md | markdown | 1.8 KB |
| references/query-rewrite.md | markdown | 1.5 KB |
| references/source-open-proof.md | markdown | 1.7 KB |
| references/subagent-attachment.md | markdown | 2.7 KB |
| references/tool-notes.md | markdown | 887 B |
| scripts/agbrowse_helper.py | script | 3.9 KB |
references/subagent-attachment.md
View on GitHub### Subagent Skill Attachment (SEARCH-ATTACH-01)
Any search subagent — Tier 3 deep-research explorers, `$cxc-lunasearch` lanes,
or ad-hoc research spawns — should receive THIS skill as a real skill
attachment, not a hand-written tool directive in the message. The subagent
auto-loads the skill at launch and follows its Tier 1/2 tool guidance
(`web_search` for discovery, then open the source for proof). The skill body is
the single source of truth for the tool list; do not duplicate it as prose in
the spawn message.
PABCD A-gate audit/reviewer dispatches are in scope too: a plan auditor must
verify references and external/current claims, so the audit dispatch packet
explicitly names `$codexclaw:cxc-search` alongside
`$codexclaw:cxc-dev-code-reviewer` (AUDIT-LOOP-01). The spawn wrapper's
`ROLE_BASE_SKILLS.reviewer` resolves the same pair when that builder is used.
The shared payload form is a **link-form mention in the spawn message**. On V1 the
child's first turn parses the mention and injects the full SKILL.md body. When a
V2-shaped spawn message reaches the codexclaw hook as plaintext (non-encrypted
provider/proxy paths), the hook recognizes the same mention and inlines the full body.
Plaintext V2 without an inlined body receives a `[CXC-SKILL-AFFORDANCE]` block
asking the child to self-load mentions from `<skillsDir>/<folder>/SKILL.md`.
Native ChatGPT-backend V2 sends ciphertext with a Fernet envelope. The hook
preserves structurally recognized ciphertext byte-for-byte and tells the caller
that hook-added skill text, scope instructions and prompt overrides were omitted;
it cannot attach the plaintext affordance to encrypted task bytes. Native recursion
checks and separate model/effort routing still apply. If the path is not link-safe, use the plugin-native
`$codexclaw:cxc-search` fallback instead:
```text
message: "[$cxc-search](skill://<this skill's SKILL.md absolute path>)
TASK: <lane / query family>"
```
On the v1 surface the structured `items` channel is equivalent and slightly
stronger (exact selection, no parse step) — use it when routing through the
spawn-wrapper builder:
```text
items: [
{ type: "skill", name: "cxc-search", path: "<this skill's SKILL.md absolute path>" },
{ type: "text", text: "TASK: <lane / query family>" }
]
```
(v2 `deny_unknown_fields` rejects `items`; plaintext V2 paths use the recognized
mention plus the hook-inlined body. The always-on spawn-attach hook never adds
`cxc-search` when the dispatcher omits it.)
Do not write a long inline TOOLS block in either path — the skill already says
"web_search for discovery, then open the source; snippets lie; the page is the
evidence." A subagent that cannot open pages must flag every finding as
`candidate — unverified snippet` in its return.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |