agentleFS
Sign inStart free
timmo001timmo001/notes/.agents/skills/effect-herdr/SKILL.md

effect-herdr

Skill2 starsChanged 20 days agoApache-2.0View on GitHub

Use @timmo001/effect-herdr for Effect TypeScript integrations with Herdr's local socket API. Apply when adding, migrating, debugging or reviewing SDK consumers, including agent coordination, event streams, panes and workspaces. For operating terminals or workspaces as an agent, use the herdr skill.

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    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:

    Prefer @timmo001/effect-herdr when an Effect application talks to Herdr.

    line 9
  2. Distinctive:

    Read the relevant current docs before coding:

    line 21
  3. Very specific:

    Keep typed failures such as HerdrUnsupportedProtocol visible.

    line 39
  4. Distinctive:

    Keep event streams and SSH agent leases within their owning scope.

    line 41
  5. Distinctive:

    Keep assignment decisions in the consumer and live state in Herdr.

    line 49
  6. Distinctive:

    Verify through the public SDK boundary.

    line 51

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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

Effect Herdr

Prefer @timmo001/effect-herdr when an Effect application talks to Herdr. Use its services and layers instead of hand-written socket clients or herdr CLI wrappers. It is a maintained fork of dmmulroy/herdr-ts-sdk published under this name. Use the separate herdr skill when operating panes, tabs or workspaces as an agent.

Read the current contract

  1. Inspect the consumer's manifest, lockfile, overrides and patches for the resolved SDK and Effect versions. Read the installed SDK README and declarations, and check its herdr.protocol metadata against the running server. Treat declared but absent packages as unverified.
  2. Read the relevant current docs before coding:
    • README: setup, API shape, errors and events.
    • Package manifest and exports: dependency pins, protocol and public surface.
    • Examples: executable recipes. They change live Herdr state, so read rather than run them. These links follow development. Prefer the installed release's docs and declarations when they differ, and make any required upgrade explicit.

Use the package

  1. Compose herdrSdkLayer (or herdrSdkLayerFromOptions) or a namespace service layer at the application boundary, and keep runtime execution there. Apply the effect skill alongside the SDK docs.
  2. Preserve boundary semantics:
    • Keep typed failures such as HerdrUnsupportedProtocol visible; install matching Herdr and SDK releases rather than bypassing the check.
    • Keep event streams and SSH agent leases within their owning scope.
    • A prompt timeout can follow delivery; inspect state before retrying.
  3. For coordination, use the agent prompt/wait operations for one-shot work and event streams for long-lived observation. Do not build polling loops or a second lifecycle state machine around them.
    • Subscriptions are live-only. For a cached view, follow Herdr's subscribe-before-snapshot ordering and resynchronise after reconnecting.
    • Keep assignment decisions in the consumer and live state in Herdr. Metadata tokens are transient display data, not durable task records.
  4. Verify through the public SDK boundary. Run the consumer's typecheck and relevant tests against its resolved packages. For live checks, target the intended Herdr server explicitly and confirm the changed operation and its cleanup. Report checked versions and any verification that was blocked.

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.

Reviews can't be read right now.

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