Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/arch3rpro/ark-space/web-searchnpx skills add arch3rPro/ark-space --skill web-searchgit clone --depth 1 https://github.com/arch3rPro/ark-spaceWrote 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/arch3rpro/ark-space/web-search)<a href="https://agentmods.dev/skills/arch3rpro/ark-space/web-search"><img src="https://agentmods.dev/badge/skills/arch3rpro/ark-space/web-search.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00074 | $0.01838 |
| Opus 5 | $0.00037 | $0.00919 |
| Sonnet 5 | $0.00015 | $0.00368 |
| Haiku 4.5 | $0.00007 | $0.00184 |
Grade A, and why
web-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 5d 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 — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Web Search
Discover public sources through the configured provider registry. This is the single query-search entry point; provider-specific syntax, filters, and troubleshooting live in docs/.
Select one explicit mode:
search: accept a query and useregistry/search-providers.yaml(Exa MCP, Exa, Tavily, Firecrawl, Jina, DuckDuckGo, Brave, SearXNG, arXiv).related: accept a seed URL and useregistry/related-page-providers.yaml(Exa similar links).
Do not infer related from an arbitrary URL: use it only when the URL is the semantic seed. Do not silently substitute a user-requested provider. Report the actual provider, request ID when supplied, result URLs, and any fallback.
Before Use
Resolve the installed ArkSpace package root before running commands. Replace <installed-arkspace-path> with the directory two levels above this loaded SKILL.md, such as /Users/<user>/.claude/plugins/cache/ark-space/ark-space/0.2.0. Use the installed package path, not a repository-relative command.
Check a provider before searching when privacy, reproducibility, or reliability matters:
python3 <installed-arkspace-path>/scripts/arkspace.py provider check exa-mcp --capability web_search
python3 <installed-arkspace-path>/scripts/arkspace.py provider check jina --capability web_search
python3 <installed-arkspace-path>/scripts/arkspace.py provider check duckduckgo --capability web_search
python3 <installed-arkspace-path>/scripts/arkspace.py provider check brave --capability web_search
python3 <installed-arkspace-path>/scripts/arkspace.py provider check tavily --capability web_search
python3 <installed-arkspace-path>/scripts/arkspace.py provider check searxng
python3 <installed-arkspace-path>/scripts/arkspace.py provider check arxiv --capability web_search
If a check fails or a provider is unconfigured, use provider-manager; never ask users to edit provider files or expose keys. arXiv needs no key or setup. SearXNG needs a self-hosted URL (provider configure searxng --base-url <url>, SEARXNG_URL, or --base-url). The keyless providers exa-mcp, jina, and duckduckgo need no configuration; brave is keyed and needs BRAVE_API_KEY.
What ships with it
16 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.
- docs/arxiv-search.md 1.9 KB
- docs/brave.md 1.3 KB
- docs/duckduckgo.md 1.2 KB
- docs/exa-mcp.md 1.7 KB
- docs/jina.md 1.1 KB
- docs/searxng-search.md 4.7 KB
- scripts/arxiv_search.py 8.8 KB runs code
- scripts/brave_search.py 8.4 KB runs code
- scripts/duckduckgo_search.py 9.7 KB runs code
- scripts/exa_mcp_search.py 20 KB runs code
- scripts/exa_search.py 14 KB runs code
- scripts/exa_similar.py 7.5 KB runs code
- scripts/firecrawl_search.py 4.6 KB runs code
- scripts/jina_search.py 7.6 KB runs code
- scripts/searxng_search.py 14 KB runs code
- scripts/tavily_search.py 10 KB runs code
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.
- 5d ago First seen · 97 lines · 74 tokens per session scan A 6d8683c8d028
web-search is a skill published in the GitHub repository arch3rPro/ark-space (2 stars, last pushed 10d ago), licensed MIT. It adds 74 tokens to every session and 1,838 once invoked, about $0.0004 per session on Opus 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 2026-08-31.
Other skills, from other repositories
invest
First-time fork users: run the invest-setup skill first to initialize. The main flow lives in this skill (the AI agent uses the CLI/MCP to view the portfolio / run the committee / replay decision history). The Web GUI has been retired (2026-07) — every capability is exposed via CLI subcommands / MCP tools. The backend…
invest-setup
First-time openInvest installation and onboarding. ONLY use when user explicitly says "set up invest" / "init invest" / "帮我初始化 invest", OR when invest skill's doctor returns status="needssetup". NOT for daily usage — once onboarding is done, the invest skill takes over (portfolio viewing, committee analysis, buy/sell…
okf-frontmatter
Maintain openInvest's docs (docs/wiki chapters + docs/wiki/adr) under Google's Open Knowledge Format (OKF). Two jobs. (1) Teach agents to maintain docs the OKF way — every doc carries a small YAML frontmatter block as the single source of truth (type, title, tags, intent, schemasource, documents); schema details link…
invest-backup
Back up / restore openInvest's local state — memory/ (holdings, strategy, user profile, committee records, dream logs) + db/ (trade ledger, job run history, market-data cache) + .env (SMTP/API credentials) + userprofile.json. All of this data is .gitignore'd with no historical versions in git, so a single accidental…
openehr-assistant
This skill should be used when a conversation touches openEHR outside a task owned by a dedicated skill — e.g. "what is an archetype?", "how do openEHR templates work?", "which composition category fits this data?", "find me a guide on X", "where do I start with openEHR modeling?" — or names openEHR concepts (ADL…
template-authoring
This skill should be used when the user asks to "create a template", "design a template", "constrain archetypes into a template", "review a template", "categorise a dataset with CGEM", "should this be persistent, episodic or event?", "split this form across compositions", "sketch a template from this form / which…