skills.rest
pjherron
pjherronCOMMUNITY
Explore all skills by this author →
GitHub

skill-invoke

Presents an Accept, Review, Revise, or Decline consent prompt before any skill executes.

Updated Jul 16, 2026
Search Tags:#agent workflow#review#skill invocation#prompt#consent#user confirmation
One-click install
npx skills add https://github.com/pjherron/hypoc --skill skill-invoke-pjherron
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-invoke
Source: https://github.com/pjherron/hypoc/tree/main/hypoc/skills/skill-invoke
Command: npx skills add https://github.com/pjherron/hypoc --skill skill-invoke-pjherron

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Automatically triggered skills can run without the user's awareness or approval. This Skill inserts a consent layer between intent detection and skill execution so every invocation is explicitly confirmed first. ## Core Features & Use Cases - Consent Prompt: Presents a plain-language description of what will happen with four options: Accept, Review, Revise, or Decline. - Review and Revise Flows: Review shows the full SKILL.md definition before running; Revise opens the definition for inline edits applied to a temporary copy, saved permanently only on explicit request. - Decline Tracking: Records declinations so the same skill is not re-suggested for the same intent during the session. - Use Case: When the platform detects that a git-workflow skill matches your request, it first asks whether you want to run it, inspect it, adjust it, or skip it, rather than firing it immediately. ## Quick Start Ask the assistant to confirm with you before running any suggested skill, showing what it will do in plain language first.

Frequently Asked Questions about skill-invoke

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

FAQPage Schema
How do I require confirmation before an AI skill runs?▼

Route every skill invocation through a consent prompt that describes the action in plain language and offers Accept, Review, Revise, and Decline options. The skill only executes after the user explicitly accepts.

How can I review a skill definition before it executes?▼

Choose the Review option in the consent prompt. The full SKILL.md content is displayed, and the prompt is presented again so you can decide whether to run it.

Can I edit a skill temporarily without saving changes?▼

Yes. The Revise option opens the SKILL.md for inline edits applied to a temporary copy, which is then run. Changes are saved permanently only if you explicitly request it.

What happens when I decline a suggested skill?▼

The declination is recorded and the same skill is not re-suggested for the same intent during the current session. Explicit slash commands bypass the consent prompt entirely.

When does the consent prompt not appear?▼

Explicit slash commands bypass the consent layer entirely, since typing the command directly is already an unambiguous expression of intent to run that skill.

Related Skills

Explore More Software Engineering Skills→
OnlyaguestOnlyaguest
18

yao-prompt-router

Route 102 Chinese AI prompts to the best-matching template.

Community
Advanced
odylithodylith
34

odylith-session-context

Preserve and resume Odylith session context across turns.

Official
Intermediate
davekilleendavekilleen
461

dex-level-up

Analyze usage logs and role profiles to suggest underutilized Dex features.

Community
Intermediate
TheCardGoatTheCardGoat
14

source-command-plan-product

Run the migrated plan-product command to plan a product and install Agent OS.

Official
Intermediate