bypass_websearch

A rule that disables the built-in web-search tool and requires web searches to use DuckDuckGo's lightweight search page instead.

In plain words
What is it for?
Use it whenever internet research is needed in a project that follows this rule. Queries must be URL-encoded, fetched from the specified page, and parsed from its returned content.
Why use it?
It ensures web lookups follow the project's chosen search route and avoids using the excluded search tool.

Cursor rule for Cursor

Install

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.

agentmods
npx agentmods add rules/nohype-ai/macstack/bypass_websearch
Clone the repo
git clone --depth 1 https://github.com/nohype-ai/MacStack

Made for: Cursor.

Per session 73 This file is loaded in full into every session.
When invoked 73 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00073 $0.00073
Opus 5 $0.00036 $0.00036
Sonnet 5 $0.00015 $0.00015
Haiku 4.5 $0.00007 $0.00007

Measured 2d ago against content hash c1b2c0c62895, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

bypass_websearch 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 2d 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.

.cursor/rules/bypass_websearch.mdc · 7 lines

What it actually says

Never use the built-in WebSearch tool. For any web search, use WebFetch on https://lite.duckduckgo.com/lite/?q=ENCODED_QUERY. URL-encode the query. Parse the results from the returned HTML/markdown.

Changes

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.

  1. 2d ago First seen · 7 lines · 73 tokens per session scan A c1b2c0c62895

Subscribe to this mod's changes

bypass_websearch is a cursor rule published in the GitHub repository nohype-ai/MacStack (6 stars, last pushed 23d ago), licensed MIT. It adds 73 tokens to every session, 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.