agentleFS
Sign inStart free
jahongashtamcjahongashtamc/hallwright/.agents/skills/subagent-driven-development/SKILL.md

subagent-driven-development

Skill1 starsChanged 13 days agoMITView on GitHub

Deconstructs complex architectures into independent tasks delegated to autonomous subagent workers, reviewing and merging each output.

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 179 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When tackling large codebases or complex multi-file features, divide and conquer by delegating bounded sub-tasks to isolated subagents …

    line 8
  2. Very specific:

    Every subagent must synchronize against the shared blackboard (.hallwright/swarm_blackboard.json or SwarmBlackboard).

    line 13
  3. Distinctive:

    No subagent may invent arbitrary data schemas, endpoints, or UI IDs.

    line 17

“Common” is measured: kept by at least 6 in 10 of the 179 edited versions of this file we could read. “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

🤖 Subagent-Driven Development

When tackling large codebases or complex multi-file features, divide and conquer by delegating bounded sub-tasks to isolated subagents while maintaining strict quality gates.

Execution Rules:

  1. Shared Blackboard State Synchronization:
    • Every subagent must synchronize against the shared blackboard (.hallwright/swarm_blackboard.json or SwarmBlackboard).
    • Ingest current project state, active contracts, and predecessor handoffs before touching code to eliminate information silos.
  2. Contract-First Protocol:
    • No subagent may invent arbitrary data schemas, endpoints, or UI IDs.
    • The lead architect registers explicit interface contracts (JSON Schemas, TypeScript interfaces, CSS token sets). Downstream subagents must strictly adhere to these registered contracts.
  3. Structured Handoff Ledger:
    • Upon stage completion, each subagent must record a formal handoff ledger entry:
      • files_modified: Exact list of files created or updated.
      • contract_ids: Contracts implemented or verified.
      • constraints_for_next: Invariants the next agent must respect.
      • guarantees: Verified test assertions and security baselines.
  4. Marcus Sterling Conflict Arbitration:
    • Intersubagent disputes (e.g. Security vs Designer on dynamic canvas, or Database vs Performance on caching) are arbitrated deterministically by Marcus Sterling.
    • Security invariants (zero XSS, zero SQLi) are non-negotiable, but Security must provide safe local primitives rather than blocking UX capabilities.
    • Once arbitrated, the ruling is bound in the blackboard state.
  5. Context Isolation & Atomic Merging:
    • Provide each subagent with only necessary bounded files to avoid token bloat.
    • Adversarially verify diffs against registered contracts and merge validated contributions atomically.

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