Hermes Agent is an AI assistant that learns from its use by creating and improving skills, retaining knowledge, searching past conversations, and adapting to its users. It is for people who want to run an agent through a terminal or messaging platforms while connecting it to different AI models and scheduled tasks.
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 NousResearch/hermes-agent --skill first-taskgit clone --depth 1 https://github.com/NousResearch/hermes-agentWrote 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/nousresearch/hermes-agent/first-task)<a href="https://agentmods.dev/skills/nousresearch/hermes-agent/first-task"><img src="https://agentmods.dev/badge/skills/nousresearch/hermes-agent/first-task/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/nousresearch/hermes-agent/first-task"><img src="https://agentmods.dev/badge/skills/nousresearch/hermes-agent/first-task.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 | $0.00013 | $0.02473 |
| Opus 5.5 | $0.00005 | $0.00989 |
| Sonnet 5.5 | $0.00003 | $0.00495 |
| Haiku 4.5 | $0.00001 | $0.00247 |
Grade A, and why
first-task 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 8d 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 — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.
First Task Skill
Runs the first chat after setup. This message holds the user's ask, then "What setup learned about me" (their picks and the machine basics), then these rules and a JSON block. Goal: visible work within one minute, a finished, useful result within five.
When to Use
- This message carries the skill under a handoff from setup.
- Until the first result lands. After that it is a normal chat.
Prerequisites
Tools: manage_connections, manage_catalog, clarify, terminal, the file tools, the browser, tool_search, skill_view, and desktop_preview in the desktop app. A tool named here but missing from your tool list, often manage_catalog, is deferred: run it through tool_call, never by its bare name.
How to Run
Obey these limits. They win over every other line.
- Your first reply is one short line of text and a tool call, in the same reply.
- The JSON block holds
connect(connector ids) andinstall(plugin ids). The ids are exact. Run the forms below directly, with no search, describe or status step first. - A check is any read-only command (
ls,which,find,cat, a probe) or a file read. At most 2 checks before you make something. Put several checks in one command. - Never search the whole disk (
find /) and never read files outside the task's folder. "What setup learned" is the survey. - At most one question before work, and only for a vague ask.
- Never run the same call twice. Change a failed call once. If it fails again, say so and go on.
- Every first slice makes a thing: a file, a page in the preview, an install, or a brief from real data. A chat summary of what you looked at is not a result.
- Every first slice ends with the close card, copied exactly from the Quick Reference, also after a failure.
- Say plainly what failed and what you did not test. Never mock or invent data.
Quick Reference
| Step | What | Call |
|---|---|---|
| 1 | One line on what you start with | none |
| 2 | Connect the picked apps | manage_connections connect, every id in connect, once |
| 3 | Install the picked plugins | manage_catalog install (through tool_call when deferred), every id in install, once |
| 4 | Specific ask: start it. Vague ask: one card, three options | clarify |
| 5 | The first slice, finished in five minutes | the task's own tools |
| 6 | One line on the result, then the close card | clarify, close card below |
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.
- 8d ago First seen · 141 lines · 13 tokens per session scan A 71a84e988d13
first-task is a skill published in the GitHub repository NousResearch/hermes-agent (254,120 stars, last pushed today), licensed MIT. It adds 13 tokens to every session and 2,473 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 2026-10-08.
Other skills, from other repositories
content-hash-cache-pattern
Cache expensive file processing results using SHA-256 content hashes — path-independent, auto-invalidating, with service layer separation.
keybindings-help
Use when the user wants to customize keyboard shortcuts, rebind keys, add chord bindings, or modify /.claude/keybindings.json. Examples: "rebind ctrl+s", "add a chord shortcut", "change the submit key", "customize keybindings".
manage-settings
Any explicit Muse Code setting question or change (model, reasoning effort, /settings) requires a silent readskill call for bundled:manage-settings as FIRST ACTION—no assistant text or other tool first. Never use for repository, app, eval, or other non-Muse configuration.
Read or triage email, clean up an inbox, draft or send messages, and check delivery. Use for the user's connected mailbox or your own email, including questions about its availability; choose the correct sender and follow the confirmation policy.
read-session
Locate and read Muse Code's OWN session logs — the current session or a prior one. Use when the user asks to pull context from, continue, summarize, or inspect a previous Muse Code session, asks to restore or recover work that was lost, wiped, or overwritten and might survive in an earlier session's log, references an…
device_data
Read cached contacts and calendar events from Muse storage. Delete Muse's local copy of either source without modifying paired devices.