Destructive Action Guard: Agent Blast-Radius Gate
Mandatory pre-execution safety gate that halts execution, computes exact blast radius, and requires typed target confirmation before high-risk Git, cloud, or system commands.
$12
Works with the AI Agents you already use
Destructive Action Guard: Agent Blast-Radius Gate
Example session with this skill installed
Agent attempts: 'gh repo delete Vikyek/production-api --yes'
- Read your context and instructions
- Compiled the destructive action guard
$ # 1. AI agent attempts to run a high-risk destructive action:
Agent: Executing gh repo delete Vikyek/production-api --yes
$ # 2. Destructive Action Guard intercepts the command before execution:
[GUARD TRIGGERED] High-risk destructive action intercepted!
Command: gh repo delete Vikyek/production-api --yes
Blast Radius Calculation
- Target: Remote GitHub Repository (Vikyek/production-api)
- Consequence: IRREVERSIBLE DESTRUCTION OF ALL CODE, COMMITS, PRS, AND ISSUES
- Recovery Path: None (Permanent data loss)
[HALT] Autonomous execution BLOCKED.
Prompt to User
"Destructive Action Guard has intercepted an irreversible repository deletion.
To execute this action, type the exact repository name: 'Vikyek/production-api' to confirm."
$ # 3. Attempting to bypass with vague confirmation fails:
User: "yes"
[GUARD] Confirmation rejected. "yes" does not match exact target name.
$ # 4. Exact confirmation provided:
User: "Vikyek/production-api"
[GUARD] Target verified. Executing command with explicit audit log entry.
[AUDIT] Action completed at 2026-11-15 20:30:00 UTC by verified confirmation.
Connects securely to your tools. The creator never sees your data.
About this skill
Agensi Turnkey Convenience Package ($12): While the core codebase remains open-source MIT on GitHub, this official Agensi package delivers the production-hardened pre-execution safety gate, verified
install.shsetup, curatedSKILL.mdagent instructions, ongoing agent compatibility updates, and priority support to protect your repositories, infrastructure, and filesystems against catastrophic autonomous agent commands.
Prevent Catastrophic AI Agent Command Execution
Autonomous coding agents given terminal access can accidentally trigger irreversible data loss (git push --force, rm -rf, gh repo delete, dropping production databases, or removing core system packages).
Destructive Action Guard acts as an unbreakable pre-execution gate. It intercepts destructive commands, calculates their exact blast radius (impacted branches, file counts, remote repositories), and halts execution until the human user provides explicit, typed confirmation.
Core Capabilities:
- Unbreakable Pre-Execution Halt: Intercepts irreversible commands before they touch the operating system or remote providers.
- Exact Blast-Radius Reporting: Computes targets, permanent consequences, and affected files/branches/repos.
- Typed Confirmation Requirement: Rejects vague "yes", "sure", or "proceed" confirmations; strictly mandates typing the exact target resource name.
- Immutable Audit Trail: Logs all verified destructive actions with UTC timestamp and operator confirmation.
See It in Action (Demo)
$ # 1. AI agent attempts to run a high-risk destructive action:
Agent: Executing `gh repo delete Vikyek/production-api --yes`
$ # 2. Destructive Action Guard intercepts the command before execution:
[GUARD TRIGGERED] High-risk destructive action intercepted!
Command: `gh repo delete Vikyek/production-api --yes`
Blast Radius Calculation:
- Target: Remote GitHub Repository (Vikyek/production-api)
- Consequence: IRREVERSIBLE DESTRUCTION OF ALL CODE, COMMITS, PRS, AND ISSUES
- Recovery Path: None (Permanent data loss)
[HALT] Autonomous execution BLOCKED.
Prompt to User:
"Destructive Action Guard has intercepted an irreversible repository deletion.
To execute this action, type the exact repository name: 'Vikyek/production-api' to confirm."
$ # 3. Attempting to bypass with vague confirmation fails:
User: "yes"
[GUARD] Confirmation rejected. "yes" does not match exact target name.
$ # 4. Exact confirmation provided:
User: "Vikyek/production-api"
[GUARD] Target verified. Executing command with explicit audit log entry.
[AUDIT] Action completed at 2026-11-15 20:30:00 UTC by verified confirmation.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed today
Requires
Permissions
File Scopes
Monitors and intercepts destructive terminal commands. Does not transmit external data.