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 skills add thomasbek3/hermes-bot-kit --skill agent-cardsgit clone --depth 1 https://github.com/thomasbek3/hermes-bot-kitWrote 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/thomasbek3/hermes-bot-kit/agent-cards)<a href="https://agentmods.dev/skills/thomasbek3/hermes-bot-kit/agent-cards"><img src="https://agentmods.dev/badge/skills/thomasbek3/hermes-bot-kit/agent-cards/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/thomasbek3/hermes-bot-kit/agent-cards"><img src="https://agentmods.dev/badge/skills/thomasbek3/hermes-bot-kit/agent-cards.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00030 | $0.02313 |
| Opus 5.5 | $0.00012 | $0.00925 |
| Sonnet 5.5 | $0.00006 | $0.00463 |
| Haiku 4.5 | $0.00003 | $0.00231 |
Grade A, and why
agent-cards 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 yesterday.
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 — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Cards
In a Hermes Desktop chat you can show a card instead of describing things in prose: a question with buttons, an approval to confirm before you act, or a task whose progress updates in place. Cards are drawn by Hermes Desktop with the Grokbot Override plugin. Slack, Telegram, the CLI and other channels do not draw them: there, write plain text.
When to show a card
- A choice or a yes/no confirmation: ask with the built-in
clarifytool (it shows a question card). - Before anything outward or irreversible (sending, posting, paying, booking, deleting): an approval card, then wait.
- A job that takes more than about a minute: one task card, kept up to date.
- Everything else: plain text. Cards only in the Hermes Desktop chat.
1. Question cards: the clarify tool
You need no files for these. Call clarify with one to five questions:
{ "questions": [ { "question": "Which night works?", "choices": ["Thursday", "Friday", "Saturday"] } ] }
- Up to four
choicesper question (each a short label). The card always adds a row where the user can type their own answer. "multi_select": truelets the user pick several.- Leave out
choicesto ask an open question. - Put your suggestion first and the card marks it recommended.
The answer comes back as the tool's result. Continue from it; do not ask the same thing again in text.
2. Approval, task and email cards: a card file plus one line
These cards live in a JSON file in your working directory for this chat. Your reply only points at it.
The card file
Write the file with your file tool at a relative path, by convention cards/<id>.json. The id is
c- plus a short slug, unique in this chat (c-7f3a). The whole file must stay under 32 KB.
{
"schema": "card.v1",
"kind": "approval",
"title": "Book the lake cabin",
"status": { "state": "awaiting-approval" },
"body": [
{ "type": "kv", "rows": [["Night", "Friday"], ["Guests", "2"], ["Total", "$180"]] }
],
"actions": [
{ "label": "Book it", "style": "primary", "send": "APPROVE c-9b book the cabin" },
{ "label": "Not yet", "send": "HOLD c-9b" }
]
}
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.
- yesterday First seen · 185 lines · 30 tokens per session scan A 2db11c993d00
agent-cards is a skill published in the GitHub repository thomasbek3/hermes-bot-kit (222 stars, last pushed today), licensed MIT. It adds 30 tokens to every session and 2,313 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-11-11.
agent-cards compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| agent-cardsthomasbek3/hermes-bot-kit | Skill | 222 | today | MIT | A | 30 | Claude Code, Codex |
| opendeskvitalops/opendesk | Skill | 92 | 23d ago | MIT | A | 77 | Claude Code |
| email-inbox-triagecyborg-garden/hermes-agent-mt | Skill | 14 | today | MIT | A | 19 | Claude Code, Codex |
| siyuancyborg-garden/hermes-agent-mt | Skill | 14 | today | MIT | A | 17 | Claude Code, Codex |
| agentmailcyborg-garden/hermes-agent-mt | Skill | 14 | today | MIT | A | 15 | Claude Code, Codex |
| google-workspacecyborg-garden/hermes-agent-mt | Skill | 14 | today | MIT | A | 21 | Claude Code, Codex |
| teams-meeting-pipelinecyborg-garden/hermes-agent-mt | Skill | 14 | today | MIT | A | 16 | Claude Code, Codex |