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 instructions/emilsvennesson/opencode-websearch/agents-mdgit clone --depth 1 https://github.com/emilsvennesson/opencode-websearchWhat 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 | $0.01690 | $0.01690 |
| Opus 5 | $0.00845 | $0.00845 |
| Sonnet 5 | $0.00338 | $0.00338 |
| Haiku 4.5 | $0.00169 | $0.00169 |
Grade A, and why
opencode-websearch AGENTS.md 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.
How it starts
The opening of the file, as written. The whole thing — 162 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
OpenCode plugin that provides web search via provider web search APIs. Supports multiple providers (Anthropic, OpenAI, GitHub Copilot, Moonshot, xAI). Multi-file TypeScript project built with Bun, linted with oxlint.
Commands
bun install # install dependencies
bun run format # oxfmt (auto-format source files in place)
bun run format:check # oxfmt --check (verify formatting, no changes)
bun run lint # oxlint (with tsconfig, unicorn/typescript/import/oxc plugins)
bun run lint:fix # oxlint --fix (auto-fix lint issues)
bun run test # bun:test unit tests
bun run typecheck # tsc --noEmit
bun run check # format:check + lint + typecheck + tests (the full quality gate)
bun run build # bun build (ESM bundle) + tsc (declaration files) → dist/
Tests use bun:test (built into Bun).
Project structure
src/
index.ts # plugin entry point — exports the plugin, wires tools
types.ts # shared types (config, provider resolution, search result shapes)
helpers.ts # generic utilities (date formatting)
config.ts # config resolution (opencode.json, env fallback)
providers/
index.ts # provider dispatch map (execute + error format)
registry.ts # provider metadata + provider detection helpers
anthropic/
index.ts # Anthropic web search (client, execution, response formatting)
openai/
index.ts # OpenAI web search (client, execution, response formatting)
xai/
auth.ts # xAI OAuth resolution, refresh, and authenticated fetch
index.ts # xAI native web search and citation formatting
copilot/
auth.ts # Copilot auth state and credential resolution
constants.ts # Copilot-specific constants
index.ts # Copilot web search (client, execution, response formatting)
shared/
errors.ts # shared provider error fallback formatting
openai-compatible.ts # shared OpenAI/Copilot client + response helpers
search.ts # shared search request/response constants and helpers
dist/ # build output (gitignored)
test/ # bun:test unit tests
.oxlintrc.json # oxlint configuration
tsconfig.json # TypeScript config (strict, ESNext)
package.json # Bun-based project, ESM module
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.
- 2d ago First seen · 162 lines · 1,690 tokens per session scan A 03bd7062a2b5
opencode-websearch AGENTS.md is an instructions file published in the GitHub repository emilsvennesson/opencode-websearch (53 stars, last pushed 9d ago), licensed MIT. It adds 1,690 tokens to every session, about $0.0085 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 2026-08-30.
Other instructions, from other repositories
openevolve CLAUDE.md
Instructions for algorithmicsuperintelligence/openevolve, covering claude.md, essential commands, development setup, install in development mode with all dependencies and or use makefile.
seektty AGENTS.md
Instructions for Hilbert-beinghappy/seektty: This repository ships one out-of-tree DeepSeek Harness Bundle. Harness remains the only owner of Agent, Session, model, settings, permissions, Profile, plugin, and persistence state.
amp-acp AGENTS.md
Instructions for tao12345666333/amp-acp, covering agents.md, commands, architecture and code style.
roamcode AGENTS.md
Instructions for burakgon/roamcode, covering repository instructions for coding agents, public-repository safety, stable release and ota contract, releasing a stable version and ota changes.
dev-workspace AGENTS.md
Instructions for iwe-org/dev-workspace, covering agent operating manual, start of every session, the operating loop, conventions and iwe basics.
ob-1 AGENTS.md
Instructions for Overbrilliant/ob-1, covering agents.md, project index, stack, commands and key files.