agentleFS
Sign inStart free
noahcalibusonoahcalibuso/herdr-fusion/skills/fusion/SKILL.md

fusion

Skill1 starsChanged 4 months agoMITView on GitHub

Fan one prompt out to multiple CLI agents in side-by-side herdr panes, then converge them into one fused answer. Use when the user wants a fused/converged multi-model answer, or asks for fusion / fuse.

5 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions5 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 5

    1. krisrice/oracle-db-skills★ 918same name
    2. oracle/skills★ 918same name
    3. Zhh9126/oracle-skills★ 4same name
    4. equinor/fusion-skills★ 2same name
    5. TheOnlyFusionCube/devin-plugin-cc★ 2same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repositoryOne other file it gives its agents.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 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.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do not paste secrets into it.

    line 24
  2. Very specific:

    Use whatever background/async shell mechanism this harness provides.

    line 32
  3. Distinctive:

    Do not close the panes or the tab the run created — the user reads them.

    line 48
  4. Very specific:

    If the run hard-fails (exit 1), read manifest.json and each worker's pane tail recorded there before diagnosing further.

    line 49

≈ “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

fusion

Drive the herdr-fusion fuse CLI from this agent session, then bring the result back here.

Works from any coding-agent harness that can run shell commands and read files (Claude Code, Cursor, Codex, PI, OpenCode, …). The harness is only the launcher — workers run in herdr panes.

Steps

  1. Preflight.
    • command -v herdr-fusion — if missing, tell the user to install with uv tool install git+https://github.com/noahcalibuso/herdr-fusion (or run once via uvx --from git+https://github.com/noahcalibuso/herdr-fusion herdr-fusion …) and stop.
    • Needs a live herdr session; if herdr status --json shows no running server, say so and stop.
  2. Compose the prompt. Workers get NO conversation context — write a fully self-contained request: the task, relevant paths in this repo, constraints, and the expected deliverable. Do not paste secrets into it.

  3. Launch in the background (a run takes 5–20+ minutes; never block the foreground on it):

    herdr-fusion fuse "<composed prompt>" [--instruction "<merge instruction>"] [--workers a,b,c]

    Use whatever background/async shell mechanism this harness provides. --session defaults to $HERDR_SESSION automatically when inside herdr. Tell the user the run has started and that the side-by-side panes are visible in a new tab of their herdr session.

  4. On completion, the final stdout line is RESULT <run-dir>. Read from that directory:
    • fused.md — the deliverable
    • manifest.json — per-worker status/timing; report any worker that failed or timed out Exit code 2 means partial: a worker died but a result was still produced.
  5. Report back: summarize the fused answer, including the consensus/divergence section, and give the run-dir path for the full artifacts.

Notes

  • Config lives in ./fusion.toml or ~/.config/herdr-fusion/config.toml (workers = slot name + CLI command; model/effort are that CLI's own flags).
  • Do not close the panes or the tab the run created — the user reads them; they clean up.
  • If the run hard-fails (exit 1), read manifest.json and each worker's pane tail recorded there before diagnosing further.
  • For a side-by-side comparison with no merge, use the opinion skill instead.

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