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 skills/darwin-agent/car-bench-trace/send-emailnpx skills add Darwin-Agent/Car-bench-TRACE --skill send-emailgit clone --depth 1 https://github.com/Darwin-Agent/Car-bench-TRACEWrote 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/skills/darwin-agent/car-bench-trace/send-email)<a href="https://agentmods.dev/skills/darwin-agent/car-bench-trace/send-email"><img src="https://agentmods.dev/badge/skills/darwin-agent/car-bench-trace/send-email.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 | $0.00000 | $0.03651 |
| Opus 5 | $0.00000 | $0.01826 |
| Sonnet 5 | $0.00000 | $0.00730 |
| Haiku 4.5 | $0.00000 | $0.00365 |
Grade A, and why
send-email 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 — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Compose and send an email (only after confirmation)
The user wants an email sent — contact details to someone, an EV trip plan, a weather update for a meeting. You draft it, confirm, and send only on an explicit "yes". The recipient's address comes from a real resolved contact, never invented, and when more than one contact could be the recipient you ask which one. The body contains only what the user asked for or what has been settled with them — you do not invent the trip's decisions, charging plan, or numbers on their behalf. The one variant that adds a tool here is a meeting-weather update, where the body needs a weather read. Resolving the recipient/contacts is the contacts-lookup skill; this skill composes and sends.
When this applies
"Email the details", "send the trip plan", "email someone from my the forecast for where we're meeting." The deliverable is a sent email; the recipient and any looked-up content have already been (or are being) resolved by contacts-lookup.
Tools
send_email({email_addresses, content_message})— send the message. Recipient address(es) come from the resolved contact's own details. The send validates each address against saved contacts: an address the user types that isn't on a contact can be rejected, and you have no tool to add or edit contacts — so don't promise to "add it" and don't fabricate. May be the removed tool.get_user_preferences({preference_categories:{productivity_and_communication:{email:true}}})— read the user's standing email preferences. Check this before sending: a stored preference can name a person to always copy (such as a secretary) or otherwise shape the recipient list. A standing recipient lives here, not in the request, so it only surfaces if you read it.get_weather({month, day, time_hour_24hformat, location_or_poi_id})— read the weather for a meeting's place around the meeting time, when the body is a weather update.get_location_id_by_location_name({location})— resolve a meeting's location to an id so the weather query can run.
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 · 82 lines · 0 tokens per session scan A 77ad4455e016
send-email is a skill published in the GitHub repository Darwin-Agent/Car-bench-TRACE (7 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,651 tokens. 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 skills, from other repositories
orbit-notion
Open Orbit briefing skill — selected by the Orbit pipeline when Notion is the user's only connected connector, or when the user explicitly scopes their daily digest to Notion. Pulls the past 24 hours of document edits, comments, mentions, and database row changes from the user's authenticated Notion connection and…
Cortex
Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
peekaboo
Capture and automate macOS UI with the Peekaboo CLI.
mochi-remind
Handle due reminders — notify the user with natural language and mark them done.