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/stufently/bing-webmaster-mcp/agents-mdgit clone --depth 1 https://github.com/stufently/bing-webmaster-mcpWrote 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/instructions/stufently/bing-webmaster-mcp/agents-md)<a href="https://agentmods.dev/instructions/stufently/bing-webmaster-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/stufently/bing-webmaster-mcp/agents-md.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 5.1 | $0.00779 | $0.00779 |
| Opus 5 | $0.00390 | $0.00390 |
| Sonnet 5 | $0.00156 | $0.00156 |
| Haiku 4.5 | $0.00078 | $0.00078 |
Grade A, and why
bing-webmaster-mcp 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 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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Working agreements for coding agents
Read SPEC.md first — it is the contract. This file is how to work in
the repo, not what to build.
Ground rules
- Do not invent API details. Method names, parameter shapes, quota numbers and
response formats come from a fetched Microsoft page or from
docs/api-surface.md, never from memory.SPEC.mdmarks every fact as verified or unverified; if you need something marked unverified, fetch it and update the marking in the same commit. - Never hardcode a quota. Read it from the API (
SPEC.md§9). - No network in tests. A test that reaches the internet is a broken test.
- No secrets in the repo. The API key comes from the environment. Never commit
a real key, a key-file name, or a site list.
.envstays ignored. - Every write goes through the apply boundary.
BING_WM_ALLOW_WRITESpicks whether a human sits between plan and apply, not whether the boundary exists. A mutating operation that reaches Bing without creating a plan record and passing throughapply_planis a bug, however it was invoked (SPEC.md§6). - Never add an apply or reject tool to the MCP server, in either mode. There is a test asserting their absence. If that test fails, the fix is to remove the tool, not to update the test. Enabling direct writes replaces the planning tools; it never adds a tool that can confirm a plan somebody else wrote.
Conventions
- Python ≥3.12, ruff (
line-length = 100,select = ["E","F","I","UP","B","SIM","S"]). - Dependency floors in
pyproject.toml, exact pins inconstraints.txt. Floors are lower bounds — do not raise one to the newest release just because it exists; that breaks every environment already running something older. - One
ops/module per domain. CLI and MCP both callops/, and neither holds logic of its own — if they can diverge in behaviour, the layering is wrong. - Error codes are a public contract. Adding one is fine; renaming one is a breaking change.
- Tests mirror modules 1:1.
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 · 58 lines · 779 tokens per session scan A c4e19686ff5c
bing-webmaster-mcp AGENTS.md is an instructions file published in the GitHub repository stufently/bing-webmaster-mcp (0 stars, last pushed 3d ago), licensed MIT. It adds 779 tokens to every session, about $0.0039 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 instructions, from other repositories
cyxj-groksearch CLAUDE.md
Claude Code instructions for chenyuxiaojin/cyxj-groksearch, covering claude.md, 常用命令 and 结构与坑.
ai-workflow AGENTS.md
Instructions for cunhaax/ai-workflow, covering ai workflow template, rules — non-negotiable, project overview, commands and architecture.
trackly-cli CLAUDE.md
Claude Code instructions for trackly-app/trackly-cli, covering trackly-cli, tech stack, backend production source of truth, directory structure and key commands.
vibetags com-example-config-AsyncTestConfig.instructions.md
Instructions for PIsberg/vibetags, covering copilot instructions for asynctestconfig and immutable type.
pupila CLAUDE.md
Instructions for FranRom/pupila, covering claude.md, where things live, stack, run locally and repo layout.
McpServer AGENTS.md
AGENTS.md instructions for sharpninja/McpServer, covering agent instructions, session start, rules, byrd test gate and where things live.