Skill Detail

Install role-specific agent personas with Agency Agents

Install and activate specialized AI agent personas from Agency Agents when a coding or operations session needs a focused role, workflow, and deliverable standard instead of a generic assistant prompt.

Templates & WorkflowsMulti-Framework
Templates & Workflows Multi-Framework Security Reviewed
โญ 96k GitHub stars
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill install-role-specific-agent-personas-with-agency-agents Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Tools required
Agency Agents repository, supported agent runtime (Claude Code, OpenClaw, Cursor, Gemini CLI, Aider, Windsurf, Copilot, opencode, Kimi, or Antigravity)
Install & setup
Clone the repository, run ./scripts/convert.sh to generate integration files, then run ./scripts/install.sh or ./scripts/install.sh –tool <runtime>. For Claude Code-only use, run ./scripts/install.sh –tool claude-code or copy selected category markdown files into ~/.claude/agents/.
Author
msitarzewski
Publisher
Independent Open Source
Last updated
May 11, 2026
Quick brief

Use Agency Agents to add a curated set of role-specific agent markdown files to Claude Code, OpenClaw, Cursor, Gemini CLI, Aider, Windsurf, Copilot, opencode, Kimi, or Antigravity. The operator selects a specialist such as frontend developer, backend architect, DevOps automator, security engineer, incident response commander, AI engineer, or codebase onboarding engineer, installs or copies the corresponding persona file, and invokes that role inside the agent session for a concrete deliverable.

How it works

What this skill actually does

Invoke this when the user wants the assistant to work in a disciplined specialist mode with explicit mission, workflows, success metrics, and output expectations. It is most useful for recurring work like UI implementation, architecture review, deployment automation, security review, incident coordination, rapid prototyping, or onboarding a codebase where the operator wants a repeatable persona/workflow rather than ad hoc prompting.

Scope boundary: this is not a generic agent framework or hosted product listing. The skill is limited to installing, adapting, and invoking the repository’s concrete markdown agent definitions and conversion/install scripts inside supported agent runtimes. It does not claim to automate arbitrary browser actions or replace each supported runtime; it provides a repeatable persona-selection workflow for operator-led agent sessions.