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/nylas/cli/agents-mdgit clone --depth 1 https://github.com/nylas/cliWhat 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 | $0.00909 | $0.00909 |
| Opus 5 | $0.00454 | $0.00454 |
| Sonnet 5 | $0.00182 | $0.00182 |
| Haiku 4.5 | $0.00091 | $0.00091 |
Grade A, and why
cli 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 — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Coding Agent Guidelines
For AI coding agents (Cursor, Copilot, Windsurf, Codex, etc.) working on this codebase.
Read these first — they are the source of truth:
CLAUDE.md— Working principles, critical rules, learnings from past mistakesdocs/ARCHITECTURE.md— Hexagonal architecture, project structure, package inventorydocs/DEVELOPMENT.md— Build, test, lint, and CI commands (make ci,make ci-full, etc.)docs/COMMANDS.md— CLI command reference.claude/rules/go-quality.md— Go style, imports, error handling, modern patterns.claude/rules/testing.md— Test organization, coverage targets, rate limiting
Everything below supplements those docs with quick-reference examples. If anything here conflicts with the above, the above wins.
Quick Reference: Shared Helpers
Don't create package-local wrappers — use these directly:
// CLI client
client := common.GetNylasClient()
grantID := common.GetGrantID(args)
// Client helpers (reduce boilerplate)
common.WithClient(args, func(ctx, client, grantID) (T, error) {
return client.DoSomething(ctx, grantID)
})
common.WithClientNoGrant(func(ctx, client) (T, error) {
return client.DoSomething(ctx)
})
// Output
common.PrintSuccess("Email sent successfully")
common.PrintError("Failed to send email", err)
common.FormatSize(bytes) // "1.5 MB"
common.FormatTimeAgo(time) // "2 hours ago"
common.PrintJSON(data) // Pretty-print JSON
out := common.GetOutputWriter(cmd) // --json/--yaml/--quiet
// Flags
common.AddJSONFlag(cmd, &jsonOutput) // --json
common.AddLimitFlag(cmd, &limit, 25) // --limit/-n
common.AddYesFlag(cmd, &yes) // --yes/-y
common.AddFormatFlag(cmd, &format) // --format/-f
// Validation
common.ValidateRequired("event ID", eventID)
common.ValidateRequiredFlag("--to", toEmail)
common.ValidateEmail("recipient", email)
common.ValidateURL("webhook URL", webhookURL)
common.ValidateOneOf("status", status, []string{"pending", "active"})
// HTTP (in adapters)
httputil.WriteJSON(w, http.StatusOK, data)
body, err := httputil.LimitedBody(r, maxSize)
// AI (in adapters/ai/)
ConvertMessagesToMaps(messages)
ConvertToolsOpenAIFormat(tools)
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 · 100 lines · 909 tokens per session scan A aff30011e366
cli AGENTS.md is an instructions file published in the GitHub repository nylas/cli (68 stars, last pushed 14d ago), licensed MIT. It adds 909 tokens to every session, about $0.0045 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
calendar-mcp copilot-instructions.md
Instructions for MarimerLLC/calendar-mcp, covering github copilot instructions for calendar-mcp, 📖 quick context, 🎯 project overview, 🏗️ project structure and 🛠️ technology stack.
olkcli CLAUDE.md
Instructions for rlrghb/olkcli, covering claude.md, what is this project?, quick reference, architecture and key patterns.
olkcli AGENTS.md
Instructions for rlrghb/olkcli, covering repository guidelines, project structure, build, test, and development commands, coding style & naming conventions and testing guidelines.
calendar-mcp CLAUDE.md
Instructions for MarimerLLC/calendar-mcp, covering project: calendar mcp, environment notes and kubernetes.
outlook-ews-mcp AGENTS.md
Instructions for viartemev/outlook-ews-mcp, covering agents.md, engineering rules, project shape, testing and conventions.
outlook-mcp CLAUDE.md
Instructions for mpalermiti/outlook-mcp, covering outlook mcp server, what this is, tech stack, commands and architecture.