Skip to content

Adonis0123/adonis-skills / skills/agent-dispatch

agent-dispatch

Flagged

Choose the lightest agent setup for a coding, review, drafting or design-research task: do it yourself, fan out to 2-3 workers with independent parts across the Claude Code, Codex, Grok Build, Cursor or other CLIs and accounts on this machine, run a scoped architecture loop, or send one brief to independent researchers; pick a cross-model reviewer, test budget and cleanup; track dispatched work, external waits and user decisions on a Run Board. Use when work may need delegation, or the user asks to dispatch, fan out, parallelize, run several agents or accounts, get another AI's review or independent design opinions, or asks how dispatched work is going. Chinese triggers: 派活、并行、要不要多开几个 agent、分给多个 Claude 账号 / codex / grok、找另一个 AI 审、对比优秀设计、多方意见、进度看板、派出的任务进行到哪了。Not for unrelated read-only questions, trivial edits, research with no delegation, a single review of an existing diff (agentic-review-handoff), or the machine-wide session list (sideby sessions). Inside another named workflow, only size its workers.

FileKindSize
GLOSSARY.mdmarkdown2.1 KB
SKILL.mdskill md27.8 KB
evals/README.mdmarkdown7.6 KB
evals/README.zh-CN.mdmarkdown7.5 KB
evals/bench/compare.pyscript5.3 KB
evals/bench/run_bench.pyscript12.3 KB
evals/bench/tasks.jsonconfig6.6 KB
evals/bench/test_bench.pyscript4.4 KB
evals/build_plugin_eval.pyscript11.3 KB
evals/evals.jsonconfig31.4 KB
evals/results.mdmarkdown15.8 KB
evals/test_eval_harness.pyscript3.5 KB
evals/trigger-eval-set.jsonconfig1.6 KB
fixtures/probe-full.jsonconfig1.6 KB
fixtures/probe-single.jsonconfig690 B
fixtures/run-midrun/B9.jsonlother191 B
fixtures/run-midrun/deploy-log.txtmarkdown42 B
fixtures/run-midrun/run.jsonconfig1.8 KB
fixtures/run-sessions/B9.jsonlother191 B
fixtures/run-sessions/deploy-log.txtmarkdown42 B
fixtures/run-sessions/ended.jsonlother95 B
fixtures/run-sessions/run.jsonconfig2.1 KB
fixtures/run-sessions/sessions.jsonconfig804 B
fixtures/session-index.jsonconfig401 B
fixtures/sideby-stub/sessions.jsonconfig1.5 KB
fixtures/sideby-stub/sidebyexecutablescript4.1 KB
references/accounts.mdmarkdown10.4 KB
references/context-budget.mdmarkdown2.9 KB
references/fallbacks.mdmarkdown4.7 KB
references/launch.mdmarkdown16.1 KB
references/measure.mdmarkdown3.3 KB
references/research-brief.mdmarkdown1.3 KB
references/run-board.mdmarkdown48.7 KB
references/task-spec.mdmarkdown5.0 KB
scripts/board.cssother12.6 KB
scripts/board_interaction.jsscript3.7 KB
scripts/board_markdown.jsscript2.9 KB
scripts/board_render.pyscript17.2 KB
scripts/dispatch_board.pyscript95.3 KB
scripts/isolated_case.pyscript1.6 KB
scripts/probe_agents.pyscript12.9 KB
scripts/sideby_cli.pyscript3.2 KB
scripts/sideby_sessions.pyscript3.3 KB
scripts/test_billing_and_quota.pyscript10.9 KB
scripts/test_board_interaction.jsscript2.1 KB
scripts/test_board_markdown.jsscript2.9 KB
scripts/test_board_reload_state.jsscript2.7 KB
scripts/test_bug_regressions.pyscript4.7 KB
scripts/test_dispatch_board.pyscript68.9 KB
scripts/test_harness_limits.pyscript11.7 KB
scripts/test_lane_states.pyscript7.5 KB
scripts/test_probe_agents.pyscript19.2 KB
scripts/test_review_fixes.pyscript6.7 KB
scripts/test_rounds_and_watch.pyscript7.3 KB

evals/bench/compare.py

View on GitHub
#!/usr/bin/env python3
"""Compare the arms of a run_bench.py results.jsonl: per task first, then in total (references/measure.md).

The verdict follows the rule in measure.md: the dispatch arm earns its tier only if it completes at least as
many tasks as solo, no task got worse, and its median wall time is lower. A task whose arms differ by a single
run is marked "check transcript": one run can be chance.

  python3 compare.py <out>/results.jsonl [--md report.md] [--base solo --other dispatch]
"""

from __future__ import annotations

import argparse
import json
import statistics
import sys


def load(path: str) -> list[dict]:
    with open(path, encoding="utf-8") as f:
        return [json.loads(line) for line in f if line.strip()]


def median(values: list[float]) -> float | None:
    values = [v for v in values if isinstance(v, (int, float))]
    return round(statistics.median(values), 2) if values else None


def summarize(rows: list[dict], base: str, other: str) -> dict:
    tasks = sorted({r["task"] for r in rows})
    per_task = []
    for task in tasks:
        entry = {"task": task}
        for arm in (base, other):
            mine = [r for r in rows if r["task"] == task and r["arm"] == arm]
            entry[arm] = {"passed": sum(1 for r in mine if r.get("pass")), "runs": len(mine),
                          "wall_s": median([r.get("wall_s") for r in mine]),
                          "cost_usd": median([r.get("total_known_cost_usd") for r in mine]),
                          "tiers": sorted({r.get("tier") for r in mine if r.get("tier")})}
        b, o = entry[base], entry[other]
        rate = lambda a: a["passed"] / a["runs"] if a["runs"] else None
        if rate(b) is None or rate(o) is None:
            entry["flip"] = "missing runs"
        elif rate(o) < rate(b):
            entry["flip"] = "worse"
        elif rate(o) > rate(b):
            entry["flip"] = "better"
        else:
            entry["flip"] = ""
        if entry["flip"] in ("worse", "better") and abs(b["passed"] - o["passed"]) == 1:
            entry["flip"] += " (one run: check transcript)"
        per_task.append(entry)
    totals = {}
    for arm in (base, other):
        mine = [r for r in rows if r["arm"] == arm]
        totals[arm] = {"passed": sum(1 for r in mine if r.get("pass")), "runs": len(mine),
                       "tasks_completed": sum(1 for e in per_task if e[arm]["runs"] and e[arm]["passed"] == e[arm]["runs"]),
                       "wall_s": median([r.get("wall_s") for r in mine]),
                       "cost_usd": median([r.get("total_known_cost_usd") for r in mine]),
                       "unknown_cost_runs": sum(1 for r in mine if r.get("lanes_without_cost"))}
    b, o = totals[base], totals[other]
    worse = [e["task"] for e in per_task if e["flip"].startswith("worse")]
    faster = b["wall_s"] is not None and o["wall_s"] is not None and o["wall_s"] < b["wall_s"]
    holds = o["tasks_completed"] >= b["tasks_completed"] and not worse and faster
    reasons = []
    if o["tasks_completed"] < b["tasks_completed"]:
        reasons.append(f"{other} completed fewer tasks ({o['tasks_completed']} vs {b['tasks_completed']})")
    if worse:
        reasons.append(f"worse on {', '.join(worse)}")
    if not faster:
        reasons.append(f"not faster (median {o['wall_s']}s vs {b['wall_s']}s)")
    return {"per_task": per_task, "totals": totals, "holds": holds, "reasons": reasons, "base": base, "other": other}


def render(s: dict) -> str:
    base, other = s["base"], s["other"]
    out = [f"| Task | {base} pass | {other} pass | {other} tier | {base} median s | {other} median s | {base} median $ | {other} median $ | Flip |",
           "| --- | --- | --- | --- | --- | --- | --- | --- | --- |"]
    for e in s["per_task"]:
        b, o = e[base], e[other]
        out.append(f"| {e['task']} | {b['passed']}/{b['runs']} | {o['passed']}/{o['runs']} | {', '.join(o['tiers']) or '-'} | "
                   f"{b['wall_s']} | {o['wall_s']} | {b['cost_usd']} | {o['cost_usd']} | {e['flip'] or '-'} |")
    out.append("")
    for arm in (base, other):
        t = s["totals"][arm]
        note = f"; {t['unknown_cost_runs']} run(s) had lanes without a reported cost" if t["unknown_cost_runs"] else ""
        out.append(f"- {arm}: {t['passed']}/{t['runs']} runs passed, {t['tasks_completed']} task(s) passed every run, "
                   f"median {t['wall_s']}s, median known cost ${t['cost_usd']}{note}")
    verdict = "holds" if s["holds"] else "does not hold: " + "; ".join(s["reasons"])
    out.append(f"- Verdict ({other} vs {base}): {verdict}")
    return "\n".join(out) + "\n"


def main(argv: list[str] | None = None) -> int:
    p = argparse.ArgumentParser(description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter)
    p.add_argument("results")
    p.add_argument("--base", default="solo")
    p.add_argument("--other", default="dispatch")
    p.add_argument("--md", help="also write the table to this Markdown file")
    args = p.parse_args(argv)
    rows = load(args.results)
    if not rows:
        print(f"{args.results} has no results; run run_bench.py first", file=sys.stderr)
        return 2
    text = render(summarize(rows, args.base, args.other))
    sys.stdout.write(text)
    if args.md:
        with open(args.md, "w", encoding="utf-8") as f:
            f.write(text)
    return 0


if __name__ == "__main__":
    sys.exit(main())
finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the workStars: 322kUpdated Claude CodeCodexVerifiedInstall
systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixesStars: 322kUpdated Claude CodeCodexVerifiedInstall
test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation codeStars: 322kUpdated Claude CodeCodexVerifiedInstall
skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.Stars: 198kUpdated Claude CodeCodexVerifiedInstall