agentsclimarketplace
Sign in

Parallel launch

Skill OutlineDriven/odin-claude-plugin/skills/parallel-launch

Use when the request has multiple independent sub-tasks, cross-domain research, or parallelizable file or module work.From its SKILL.md

Install
npx -y skills add OutlineDriven/odin-claude-plugin --skill parallel-launch

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

2.1 KB, 389 tokens by cl100k_base, as published. Nobody here has run it

Parallel Launch

Decompose the given task into independent agent groups and execute them in broad parallel.

Process

  1. Analyze the task and identify independent concerns that can run concurrently.

    • Each concern must be self-contained: no shared mutable state, no ordering dependency.
    • If concerns have dependencies, sequence the dependent batch after the independent batch completes.
    • Consult delegation scenarios for parallelism decisions.

  2. Design agent groups for each independent concern:

    • Assign a clear, scoped objective (one concern per agent).
    • Select the appropriate tailored agent type (Explore, Plan, or a domain specialist) — prefer a task-specific agent over a general one.
    • Define expected output format so results can be composed.

  3. Launch all independent agents in a single tool call.

  4. Compose results once all agents complete:

    • Merge non-conflicting outputs directly.
    • For conflicting or overlapping results, reconcile and present trade-offs to the user.
    • If any agent failed or returned incomplete results, report the gap and propose a targeted follow-up.

  5. Review composed output by dispatching a review agent to verify:

    • Completeness: All original concerns addressed, no gaps.
    • Consistency: No contradictions between agent outputs.
    • Accuracy: Claims are substantiated, sources checked, no hallucinated findings.
    • Scope: Nothing extra built beyond what was asked.
    • For implementation work, additionally verify spec compliance and code quality.

  6. Report to user only after review passes.

Constraints

  • Agents per batch: match the number of truly independent concerns (avoid artificial splitting).
  • Each agent prompt must include full context; agents do not share memory.
  • If the task has fewer than 2 independent concerns, execute directly instead of launching agents.

What ships with it: 1 file

1.3 KB alongside SKILL.md

references/

Gives 0 of the 12 instructions most product growth skills give in 389 tokens

Counted across 694 of the 879 authors here whose files we hold, read 2026-09-06

  • Check for product marketing context firstin 49 of 694, across 20 files
  • Validate the why before building featuresin 18 of 694, across 4 files
  • Respond to every comment in real-timein 17 of 694, across 6 files
  • Structure launch marketing across three channel typesin 16 of 694, across 4 files
  • Recruit early users one-on-onein 13 of 694, across 2 files
  • Ask one question at a timein 13 of 694
  • Rank features using ICE scoringin 12 of 694, across 3 files
  • Identify primary conversion goalin 11 of 694, across 3 files
  • Identify traffic contextin 11 of 694, across 3 files
  • Evaluate headline effectivenessin 11 of 694, across 3 files
  • Check visual hierarchy and scannabilityin 11 of 694, across 3 files
  • Run product diagnosticsin 11 of 694, across 3 files

Said here and by no other author read

  • Analyze the task to identify independent concerns
  • Design agent groups for each independent concern
  • Launch all independent agents in a single tool call
  • Compose results once all agents complete
  • Review composed output by dispatching a review agent
  • Report to user only after review passes

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.