agentleFS
phnx-labsphnx-labs/rush-mcp/skills/save-an-agent/SKILL.md

save-an-agent

Skill0 starsChanged 4 days agoMITView on GitHub

Save a reusable Rush agent with standing instructions, model and connected accounts, or add a ready-made one from Community, so repeat work starts by name. Use when someone asks for the same kind of task again and again (weekly triage, release notes, PR reviews to a team's standard), or asks to set up, change or find a Rush agent.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Check what exists: Call list_agents with a query for the job.

    line 11
  2. Very specific:

    Use and refine: Start tasks with create_session and agent set to its name.

    line 36

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Save a Rush agent

A saved agent fixes the harness, model, instructions, skills and connected accounts, so each task only says what is new this time.

  1. Check what exists. Call list_agents with a query for the job. To start from someone else's work, list_agents with community: true lists agents companies published; create_agent with only template set to its id adds it as the person's own, running on their connected accounts.
  2. Write the instructions. Turn what the person keeps repeating into standing instructions: what the agent is for, the steps it always takes, the standard its output meets, and what it must never do (for example, "never merge" or "draft replies, do not send"). Leave out anything that changes per task; that goes in each task.
  3. Choose the rest. harness is claude, codex or opencode; leave model and effort out for the harness default, or pick from list_models. Name only the accounts it needs in connections (list_connections); Gmail and Outlook take {name, access: "read"} to read without sending. Add skills as [{url}] (for example a GitHub folder URL) when the person has them. GitHub is not a connection: the repositories the Rush GitHub App reaches are in list_repositories, and each task names its repo.
  4. Save it. Show the person the name and instructions, then call create_agent. If a connection it names is not connected yet, call connect_account with type and name and give the person the authorize_url to open themselves. An agent added from a template runs only on its own connections: read what it lacks with list_agents and template set to the template (missing_connections), and connect each whose: "agent" entry with connect_account and agent set to the new agent. A whose: "template" entry only the template's owner can repair.
  5. Use and refine. Start tasks with create_session and agent set to its name. Change it with update_agent; sessions already running keep the version they started with. The person can also edit it at https://cloud.getrush.ai/console/agents/

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.