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/orieg/gws-connector/copilot-instructionsgit clone --depth 1 https://github.com/orieg/gws-connectorWrote 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/orieg/gws-connector/copilot-instructions)<a href="https://agentmods.dev/instructions/orieg/gws-connector/copilot-instructions"><img src="https://agentmods.dev/badge/instructions/orieg/gws-connector/copilot-instructions.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 | $0.00390 | $0.00390 |
| Opus 5 | $0.00195 | $0.00195 |
| Sonnet 5 | $0.00078 | $0.00078 |
| Haiku 4.5 | $0.00039 | $0.00039 |
Grade A, and why
gws-connector copilot-instructions.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 4d 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.
What it actually says
Copilot Code Review Instructions
Project Overview
gws-connector is a multi-account Google Workspace MCP server (Go). It connects Gmail, Calendar, and Drive to Claude Code and other MCP clients via JSON-RPC over stdio.
Architecture
cmd/gws-mcp/— Entry point, reads env vars, starts MCP server on stdiointernal/accounts/— Account registry (JSON file) and routing (label/email resolution)internal/auth/— OAuth2 flows, token storage (OS keychain with file fallback), API client factoryinternal/services/— Gmail, Calendar, Drive tool handlersinternal/server/— MCP tool registration and dispatch
Review Focus Areas
Security (high priority)
- OAuth token handling: tokens must never be logged or included in error messages
- File permissions: state files must use 0600, directories 0700
- Input sanitization: parameters used in Google API queries (especially Drive
folderId) must be validated - No hardcoded credentials — all secrets come from env vars or OS keychain
- HTML output in OAuth callback must use
html.EscapeString(), not manual replacement
Correctness
- Account routing: label/email matching must be case-insensitive (
strings.EqualFold) - Error handling: all errors from external calls must be wrapped with context using
%w - Token refresh: persisted tokens must be saved after refresh; save errors must be logged
Go idioms
- Use
fmt.Errorf("context: %w", err)for error wrapping - Prefer early returns over deep nesting
- No exported symbols unless needed by another package
- Tests use
t.TempDir()for filesystem isolation
Out of scope
- Performance optimization (single-user MCP server, not a web service)
- Concurrency patterns beyond the OAuth callback goroutine
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.
- 4d ago First seen · 38 lines · 390 tokens per session scan A 6530c59be0bc
gws-connector copilot-instructions.md is an instructions file published in the GitHub repository orieg/gws-connector (1 stars, last pushed 2d ago), licensed MIT. It adds 390 tokens to every session, about $0.0019 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
notfair-plugin AGENTS.md
AGENTS.md instructions for nowork-studio/notfair-plugin, covering notfair — skill resolver for ai agents, how to use this file, seo, paid ads and analytics.
notfair-plugin CLAUDE.md
Claude Code instructions for nowork-studio/notfair-plugin, covering engineering execution standard, agent entry points (single sources of truth), working style: brutal honesty, relentless quality, repository purpose and critical: this ships to users.
domscribe CLAUDE.md
Instructions for patchorbit/domscribe, covering claude.md, what is domscribe?, how to work on this codebase, commands and install dependencies (pnpm, not npm).
writing-skills AGENTS.md
AGENTS.md instructions for surendranb/writing-skills, covering agents.md — writing-skills, what this is, layout, the skill anatomy contract (enforced by ci) and working here.
logic-lens GEMINI.md
Instructions for hyhmrright/logic-lens, covering logic-lens — developer guide (gemini cli), invoking skills, installation, project layout and key conventions.
flutter-slipstream CLAUDE.md
Instructions for devoncarew/flutter-slipstream, covering flutter-slipstream, feature: flutter ui agent, feature: package api retrieval, feature: package safety skill and supported coding agents.