agentic-engineering-capabilities

Inspect, register, validate, and disable project tools and provider integrations.

Updated Jul 19, 2026
One-click install
npx skills add https://github.com/DagimAlemayehuu/AgenticEngineering --skill agentic-engineering-capabilities-dagimalemayehuu
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agentic-engineering-capabilities
Source: https://github.com/DagimAlemayehuu/AgenticEngineering/tree/main/agentic-engineering/skills/agentic-engineering-capabilities
Command: npx skills add https://github.com/DagimAlemayehuu/AgenticEngineering --skill agentic-engineering-capabilities-dagimalemayehuu

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams connecting coding agents, research workspaces, design tools, MCP services, and deployment integrations to a project often lack a safe, auditable way to verify permissions, data exposure, and readiness before enabling them, risking credential leaks or silent provider failures. ## Core Features & Use Cases - Capability Inspection: Check availability, permissions, data exposure, cost, and failure behavior of any tool or provider before registering it. - Deterministic Registry Management: Use the capabilities.py script to inspect, register, validate, and disable entries in the canonical .agentic/capabilities.json registry, which stores identifiers and readiness facts but never credentials. - Readiness Reporting: Produce an updated capability registry plus a plain-language readiness report, treating unavailable providers as visible blocked states rather than silently substituting them. - Use Case: Before wiring a new MCP service into your agent workflow, run this Skill to confirm authority and data boundaries, register the capability, and get explicit approval gates before uploading data or changing production access. ## Quick Start Use $agentic-engineering-capabilities to inspect and register the new MCP service for this project.

Frequently Asked Questions about agentic-engineering-capabilities

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

FAQPage Schema
How do I safely register a new MCP service or provider for my project?▼

Inspect the provider's availability, permissions, data exposure, cost, and failure behavior first, then register it in .agentic/capabilities.json using the capabilities.py script. Registration records identifiers and readiness facts only, never credentials or API keys.

How do I diagnose why a coding agent or provider integration is not working?▼

Run the capability inspection to check availability and permissions, then treat an unavailable provider as a visible blocked state. The workflow forbids imitating or silently substituting a failed provider, so the readiness report surfaces the real cause.

Does registering a capability prove the provider is operational?▼

No. Registration only records identifiers and readiness facts in the registry. Only a successful host probe can prove a provider is actually operational, so validation should include a live probe before relying on it.

Can the capability registry store API keys or tokens?▼

No. The canonical .agentic/capabilities.json registry may contain identifiers and capability facts but never credentials, tokens, or API keys. This boundary is enforced by the workflow to prevent credential leakage.

When does the capability workflow require explicit human approval?▼

Approval is required before uploading data, spending money, or changing production access. The inspection and registration steps stop at those boundaries and produce a readiness report for review instead of proceeding autonomously.