Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/taxueseek/argonpx agentmods add skills/taxueseek/argo/local-searchWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/taxueseek/argo/local-search)<a href="https://agentmods.dev/skills/taxueseek/argo/local-search"><img src="https://agentmods.dev/badge/skills/taxueseek/argo/local-search/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/taxueseek/argo/local-search"><img src="https://agentmods.dev/badge/skills/taxueseek/argo/local-search.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00059 | $0.01434 |
| Opus 5.5 | $0.00024 | $0.00574 |
| Sonnet 5.5 | $0.00012 | $0.00287 |
| Haiku 4.5 | $0.00006 | $0.00143 |
Grade A, and why
local-search scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 28d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Local Search 子技能
Local Search 是 argo 的「零成本兜底适配器」，用于：
- 在
--mode fast/--mode budget下优先使用本地抓取引擎，避免消耗付费 API 配额。 - 当 SearXNG 不可用时，回退到本地 HTML/JSON 解析。
- 对中文网页、新闻、代码问答、学术、参考百科等垂直域提供补充结果。
设计原则
- 不单独响应触发词：仅作为 argo 的子能力，由
--local-first调用。 - 输出与主 skill 同 schema：
results[]/engines_used/errors/elapsed_ms，可直接进 evidence 与 RRF。 local_X与主清单的X是别名，不是备份：local_arxiv/local_crossref/local_semantic_scholar/local_github/local_npm/local_wikipedia/local_stackoverflow/local_google_news与主清单同名项打同一个上游端点（主清单走官方 API、这里走免密钥直取），不存在「API 挂了抓取版兜底」。同上游同格式由tests/test_local_search_registry.py锁定。
本地引擎列表（29 个，27 个默认启用）
| unified 名称 | 类型 | 默认启用 | 类别 | 说明 |
|---|---|---|---|---|
| local_bing | cli(ddgs) | ✅ | web_general | Bing 网页结果（ddgs -b bing + JSON） |
| local_yandex | cli(ddgs) | ✅ | web_general/japanese | Yandex 搜索（ddgs -b yandex） |
| local_duckduckgo | cli(ddgs) | ✅ | web_general | DuckDuckGo（ddgs 默认） |
| local_brave | cli(ddgs) | ✅ | web_general | Brave 搜索（ddgs -b brave，实测稳定） |
| local_yahoo | cli(ddgs) | ✅ | web_general | Yahoo 搜索（ddgs -b yahoo，实测稳定） |
| local_baidu | html | ✅ | chinese | 百度搜索 |
| local_sogou | html | ✅ | chinese | 搜狗搜索 |
| local_360 | html | ✅ | chinese | 360 搜索 |
| local_jisilu | html | ✅ | finance/chinese | 集思录 |
| local_ddgs_news | cli(ddgs) | ✅ | news | ddgs news 默认后端（带日期） |
| local_bing_news | rss | ✅ | news | Bing 新闻 RSS |
| local_google_news | rss | ✅ | news | Google News RSS |
| local_duckduckgo_news | cli(ddgs) | ✅ | news | ddgs news 备用后端 |
| local_ddgs_images | cli(ddgs) | ✅ | images | ddgs images（bing TLS 偶发，已自动重试） |
| local_ddgs_videos | cli(ddgs) | ✅ | videos | ddgs videos |
| local_arxiv | xml | ✅ | academic | arXiv API |
| local_pubmed | json | ✅ | academic | PubMed/EUtils |
| local_crossref | json | ✅ | academic | Crossref API |
| local_semantic_scholar | json | ✅ | academic | Semantic Scholar API |
| local_github | json | ✅ | code | GitHub Search API |
| local_stackoverflow | json | ✅ | code | StackOverflow 问题 |
| local_gitlab | json | ✅ | code | GitLab API |
| local_npm | json | ✅ | code | NPM Registry |
| local_wikipedia | json | ✅ | reference | MediaWiki API（ddgs wikipedia backend 结果过少） |
| local_wiktionary | json | ✅ | reference | Wiktionary API |
| local_wikiquote | json | ✅ | reference | Wikiquote API |
| local_imdb | html | ❌ | vertical | IMDb 搜索 |
| local_goodreads | html | ❌ | vertical | Goodreads 搜索 |
| local_openstreetmap | json | ✅ | vertical | Nominatim API |
What ships with it
10 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 28d ago Changed · -1 lines · -7 tokens per session 5498c3cdb9ba
- 1mo ago Changed · -3 lines a33e948b1e2a
- 1mo ago Changed · -9 lines 75073b4ff4a9
- 1mo ago First seen · 115 lines · 66 tokens per session scan A 5024af890b29
local-search is a skill published in the GitHub repository taxueseek/argo (209 stars, last pushed today), licensed Apache-2.0. It adds 59 tokens to every session and 1,434 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
audit-playbook
A code-security audit playbook covering source-code review, runtime checks, software dependencies, deployment settings, and AI-agent security risks.
design-experiment
Translate a falsifiable hypothesis into a statistically defensible experiment. Factorial, ablation, comparison, scaling, and robustness designs are modes, not strategy nodes.
artifact-detection
Detect annotation artifacts and shortcuts in benchmarks.
audit-validator-independence
Audit whether a validator, benchmark, sandbox, or test can pass by construction because it embeds the same assumptions as the theory it claims to validate. Build a claim×validator-assumption matrix and demand at least one non-circular failure channel.
adjust-abstraction-scope
Move a scientific object up/down in abstraction or narrow/broaden selected scope dimensions (population, mechanism, context, outcome, timeframe, system boundary) until the representation has useful explanatory and experimental leverage.
apply-stage-gate
Apply explicit GO / HOLD / RECYCLE / KILL (or supplied) stage-gate criteria with evidence thresholds and next-step conditions.