agentleFS
Sign inStart free
p10ns11yp10ns11y/skills/agentic-reactor/SKILL.md

agentic-reactor

Skill2 starsChanged 40 days agoMITView on GitHub

Overarching patterns for self-guarded, pause-aware, agent-driven desktop/web platforms. Combines finder-reactor, Tauri shell, X resources, and CV guard into a cohesive autonomous system with intervention only on guards. Use for high-level autonomy design, meta-improvement loops, MCP composability. Fusion for the reactor as living system; fission for specific guards.

The file

We split this skill into the rules it gives an agent

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

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Self-Improvement: After cycles, surplus proposals — auto-apply low-risk, pause for high-impact.

    line 17
  2. Distinctive:

    After reactor-related work, document Q-factor wins and suggested follow-ups.

    line 36

≈ “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

Agentic Reactor — The Living, Self-Improving Finder Platform

CLT: ../rules/clt-dual-load.mdc — autonomy continues under guards; human pauses are short previews + real decisions (germane); do not spam concurrent pause dialogs.

Core Mission: Make an opportunity-finder not just a tool, but an autonomous agent that runs with minimal human input, using self-guards and pauses to stay safe. It decides, acts (search, prep), learns, and improves — surfacing only when necessary (low fit, high cost, CV changes).

Principles

  • Autonomy with Guardrails: Reactor loop runs search → analyze → decide → prep if cleared → track.
  • Pauses: Explicit intervention points via UI dialogs and MCP pause responses.
  • Smart Decisions: Structured "next action" with guards evaluated in the core (Rust or backend service).
  • Self-Improvement: After cycles, surplus proposals — auto-apply low-risk, pause for high-impact.
  • Composability: Full MCP exposure so Cursor, Grok, or other agents can invoke guarded hunts.
  • Integration: X via x-agent-resources; CV via cv-promote-guard; shell via tauri-agentic.

Architecture (Fusion View)

FinderReactor (core):

  • State: leads/preps in app data (JSON or SQLite).
  • Guards: Cost, Rate, Fit, CVPromote (delegates to cv-promote-guard).
  • X layer: x-agent-resources patterns.
  • MCP server: reactor methods + ask_user for pauses.
  • Optional background daemon with tray/UI surfacing pauses.

Shell (Tauri or web):

  • Command palette + guard dashboards + pause modals.
  • Frontend: react-client-expert minimal state.

Surplus Protocol

After reactor-related work, document Q-factor wins and suggested follow-ups. Track in surplus-log or fusion-state.

Activation

High-level design: load agentic-reactor + finder-reactor + tauri-agentic + fusion-sage.

Example provenance: collab-finder.

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