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 clutchpbcfo/hermes-skills --skill skill-writergit clone --depth 1 https://github.com/clutchpbcfo/hermes-skillsWrote 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/clutchpbcfo/hermes-skills/skill-writer)<a href="https://agentmods.dev/skills/clutchpbcfo/hermes-skills/skill-writer"><img src="https://agentmods.dev/badge/skills/clutchpbcfo/hermes-skills/skill-writer/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/clutchpbcfo/hermes-skills/skill-writer"><img src="https://agentmods.dev/badge/skills/clutchpbcfo/hermes-skills/skill-writer.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 | $0.00057 | $0.00954 |
| Opus 5 | $0.00028 | $0.00477 |
| Sonnet 5 | $0.00011 | $0.00191 |
| Haiku 4.5 | $0.00006 | $0.00095 |
Grade A, and why
skill-writer 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 11d 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 — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill Writer
When to Use
Use this when the user wants to create a new Hermes skill, clean up one they have, or turn a repeatable workflow they keep doing by hand into a saved skill. Triggers include "help me write a skill for...", "turn this workflow into a skill", "make a SKILL.md", or "I keep doing X manually, can we save it." The output is a complete, installable SKILL.md, not advice about writing one.
Background
A skill is an on-demand knowledge document the agent loads only when a task needs it (progressive disclosure). Skills follow the agentskills.io open standard: a Markdown file with YAML frontmatter, stored at ~/.hermes/skills//SKILL.md. The same format is read by other agents, so a well-written skill is portable.
Procedure
- Extract the workflow. Ask the user to describe the task they do repeatably. Pull out: the goal, the trigger (when they would want this), the steps in order, the inputs needed, and how they know it worked. If they describe it vaguely, ask for one concrete recent example.
- Name it. Choose a short, lowercase, hyphenated name that reads as what it does (for example invoice-chaser, not helper-2).
- Write the description. One sentence, written so it matches how a person would actually phrase the request. This field is what makes the agent decide to load the skill, so it must contain the trigger language, not just a label.
- Write the frontmatter. Produce valid YAML: name, description, version (start at 1.0.0), optional platforms only if the skill is OS-specific, and metadata.hermes with tags and a category.
- Write "When to Use." Spell out the trigger conditions and a few real phrasings a user might say. Vague trigger text is the number one reason a skill never loads, so make this specific.
- Write "Procedure." Numbered, ordered steps. Each step is one clear action. Define what "done" looks like. Where the order matters, say so.
- Write "Pitfalls." List the real failure modes for this workflow and how to avoid them. Skip generic warnings; name the ones that actually bite.
- Write "Verification." Give a concrete success check the agent can apply to know it finished correctly, not just "make sure it worked."
- Assemble and deliver. Output the full SKILL.md in one block, ready to save to ~/.hermes/skills//SKILL.md. Then tell the user how to test it (see Verification).
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.
- 11d ago First seen · 41 lines · 57 tokens per session scan A a1d92901c0db
skill-writer is a skill published in the GitHub repository clutchpbcfo/hermes-skills (4 stars, last pushed 3mo ago), licensed MIT. It adds 57 tokens to every session and 954 once invoked, about $0.0003 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 skills, from other repositories
mnemosyne
Persistent cross-session memory via Mnemosyne — store, recall, and consolidate facts, preferences, and context.
mnemosyne-memory-override
Hard rule override that forces Mnemosyne for all durable memory storage. The legacy memory tool is DEPRECATED for user preferences, credentials, and project conventions. Use memory ONLY for ephemeral session state.
life-os
You are Hermes Life OS — a personal operating system that learns who you are, remembers everything you share, and grows smarter about your life every single day.
hermes-tweet
Use Xquik in Hermes Agent for public X research, monitoring, thread summaries, creator discovery & approved actions. Not affiliated with X Corp. Use when the user requests X data or a named action. Trigger with "search X", "monitor X", "post tweet", or "X trends".
vrc-monitor-agent
Use for VRChat friend queries (online status, who played with whom, activity timelines, online patterns) and VRChat social actions (boop, invite, join world, friend management, group operations, image uploads) via the vrc-monitor MCP server on port 8799.
mnemosyne-maintenance
Use when: upgrading Mnemosyne, diagnosing slow/hung consolidation (mnemosynesleep), fixing missing embeddings, or troubleshooting import/version mismatches.