PuntoyComaTechPuntoyComaTech/hermes-agent-squads/skills/hermes-platform/SKILL.md

hermes-platform

Skill3 starsChanged 18 days agoMITView on GitHub

Indexes verified Hermes Agent primitives. Use when designing or changing a squad, bot or Hermes config.

In this repository11 other files it gives its agents.
- AGENTS.mdAGENTS.md
- llms.txtllms.txt
- assigning-modelsSkill
- authoring-skillsSkill
- changing-botsSkill
- contributing-upstreamSkill
- designing-squadsSkill
- finding-capabilitiesSkill
- installing-squadsSkill
- managing-projectsSkill
- tracking-updatesSkill

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Very specific:
Before writing or editing a squad.yaml, a SOUL, a skill or a profile config.
line 10
Distinctive:
Before answering how profiles, kanban, cron, plugins, MCP or gateways behave.
line 11
Distinctive:
When a squad design depends on a Hermes behavior (per-card models, Projects, delivery targets, delegation).
line 12
Very specific:
Before planning a Desktop plugin, a new pane or any UI that reads across profiles: references/desktop-plugins.md.
line 14
Very specific:
Read skills/hermes-local-notes/SKILL.md when it exists.
line 19
Very specific:
When a reference and the live Hermes (hermes <cmd> --help, the source) disagree, the live Hermes wins.
line 25

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

SKILL.md

Hermes platform

When to Use

Before writing or editing a squad.yaml, a SOUL, a skill or a profile config.
Before answering how profiles, kanban, cron, plugins, MCP or gateways behave.
When a squad design depends on a Hermes behavior (per-card models, Projects, delivery targets, delegation).
Before planning a Desktop plugin, a new pane or any UI that reads across profiles: references/desktop-plugins.md.

Procedure

Read skills/hermes-local-notes/SKILL.md when it exists; its facts win over references/.
Pick the topic from the index below and read that reference file. The references were verified against the Hermes version in scripts/data/doc-sources.json (verified_against); on a newer Hermes, check a fact you act on against --help or the source.
When a reference and the live Hermes (hermes <cmd> --help, the source) disagree, the live Hermes wins. Append <topic>: <fact> (<date>) to skills/hermes-local-notes/SKILL.md, creating it with frontmatter name: hermes-local-notes and description: "Lists local Hermes facts that override hermes-platform. Use when a reference and live Hermes disagree.". Then propose the fix upstream (skills/contributing-upstream).
For anything deeper, read website/docs/ under the Hermes source root: the Install directory: line of hermes --version.

Reference index

Topic	File
Profiles, distributions, install/update	`references/profiles-and-distributions.md`
SOUL.md, context files, ui_meta	`references/soul-and-context-files.md`
Skill format and sources	`references/skills-format-and-sources.md`
Toolsets	`references/toolsets.md`
Browser backends, real profile, signing in for bots	`references/browser.md`
Bot Mode, Desktop, Bots pane, rail	`references/bot-mode-and-desktop.md`
Bot-to-bot messaging	`references/bot-to-bot.md`
Kanban cards, workers, per-card flags	`references/kanban.md`
Cron, gates, delivery	`references/cron-and-gates.md`
Delegation and the advisor pattern	`references/delegation.md`
Models, auxiliary slots, providers, credential pools, prompt cache	`references/models-and-auxiliary.md`
Built-in memory and memory providers	`references/memory.md`
MCP servers	`references/mcp.md`
Tool search and connectors	`references/tool-search-and-connectors.md`
Plugins and the catalog	`references/plugins.md`
Desktop plugins: surfaces, host doors, limits	`references/desktop-plugins.md`
Approvals and security	`references/approvals-and-security.md`
Hooks	`references/hooks.md`
Projects	`references/projects.md`
Gateways and messaging	`references/gateways-and-messaging.md`

Pitfalls

A flag or config key from memory: re-read the reference or run --help.
hermes config get <key> warns "not a recognized config key" for valid keys outside the schema (agent.reasoning_effort): read those with hermes config show or the profile's config.yaml.
UNVERIFIED notes are hypotheses, not behavior.
A Desktop plugin surface from memory: check apps/desktop/src/sdk/index.ts; the SDK doc page omits side effects core runs.

Verification

The answer matches hermes <cmd> --help on this machine.
scripts/hpy skill_lint.py skills/hermes-platform exits 0.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free