skills.rest
dividedby
dividedbyCOMMUNITY
Explore all skills by this author →
GitHub

apply-agent-research

Audit repository governance and skills against external agent-research knowledge.

1|2|Updated May 27, 2026
Search Tags:#automation#audit#agent#repository#governance#proposal
One-click install
npx skills add https://github.com/dividedby/skills --skill apply-agent-research
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: apply-agent-research
Source: https://github.com/dividedby/skills/tree/main/skills/meta/apply-agent-research
Command: npx skills add https://github.com/dividedby/skills --skill apply-agent-research

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires json_repair, and includes scripts (resource) components.

What problem does it solve?

This skill solves the problem of keeping a repository's agent-meta configuration (instructions, hooks, and skills) aligned with evolving best practices without requiring manual intervention or risky automated commits.

Core Features & Use Cases

  • Decentralized Knowledge Consumption: Reads external agent-research knowledge bases to identify potential improvements for the host repository.
  • Safe Proposal Workflow: Audits the repository's governance docs (CONTEXT.md, ADRs) and existing skills to propose improvements exclusively via issues, ensuring no unauthorized mutations occur.
  • Adversarial Filtering: Uses a multi-stage gate to filter out redundant, vague, or low-frequency proposals, ensuring only high-leverage changes reach the maintainer.

Quick Start

Run the apply-agent-research skill to audit the current repository against the latest agent-research knowledge base and file a single, ranked improvement proposal if a high-value gap is identified.

Frequently Asked Questions about apply-agent-research

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate agent-meta configuration updates without risky direct commits?▼

This skill automates agent-meta configuration updates by proposing improvements exclusively via issues, auditing governance docs and skill inventories against a knowledge mirror to prevent unauthorized mutations.

What is a decentralized agent-meta improvement loop?▼

A decentralized agent-meta improvement loop applies external agent-research knowledge to a repository by identifying integration gaps between governance docs and public knowledge mirrors, then filing ranked improvement proposals without manual intervention.

How does adversarial filtering work for automated repository proposals?▼

Adversarial filtering for automated repository proposals uses a multi-stage gate to block redundant, vague, or low-frequency suggestions, ensuring only high-leverage changes that satisfy safe unattended operation budgets reach the maintainer.

Can I audit repository governance docs against external knowledge bases?▼

Yes, you can audit repository governance docs like CONTEXT.md and ADRs against an external public knowledge mirror to identify integration gaps and file a single ranked improvement proposal if a high-value issue is found.

Does this proposal workflow require json_repair to parse knowledge bases?▼

Yes, the proposal workflow requires the json_repair dependency to parse external agent-research knowledge bases, ensuring accurate data extraction for auditing local skill inventories and generating safe improvement proposals.

Related Skills

Explore More Software Engineering Skills→
cuiosscuioss
4

tools-marketplace-inventory

Scan marketplace components and generate TOON or JSON inventory reports.

Official
Intermediate
BoardPandasBoardPandas

doc-sync

Cross-references documentation files against public APIs, endpoints, and configuration options to generate an issue report.

Official
Intermediate
acme-skunkworksacme-skunkworks

grill-with-docs

Validate project plans against domain models and update CONTEXT.md and ADRs.

Official
Intermediate
rjwaltersrjwalters
1

anvil-proposal

Register the Anvil proposal skill for Claude via a lightweight shim.

Community
Basic