Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for penetration testers

Social Engineering Pretext Approval Agent

A pretext plan that passes every rule of engagement and has written approval from both the client and the lead

Risk managementComplianceQuality controlIT and development
See how it worksGet access
Social Engineering Pretext Approval Agent: what goes in, what the agent does and what you get

What it does

A social engineering test can go wrong in ways a technical test cannot. A pretext can scare staff, impersonate a real person, or involve a call at a bad time. This agent reads each planned pretext, target list and timing. It checks them against the rules of engagement and the legal limits the team has recorded, such as banned topics, protected groups, off-limits roles and call hours. It flags every issue with the exact rule behind it and drafts a safer version of the pretext. Then it checks the revised version against the same rules and repeats until nothing is flagged. Nothing is used until the client contact and the lead both approve. Edge case: a phishing email mentions a layoff, so the agent replaces it with a neutral delivery notice.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueYes, continueApprovedNoNo 1 STARTS WHEN Tester drafts a pretext, call plan or target list 2 USES A TOOL Read the rules of engagement and legal limits 3 DOES Break the plan into message, target group, senderidentity and timing 4 CHECKS THE RESULT Does every part pass the rules of engagement andlegal limits? If not: list each broken rule and draft a safer versionof that part. Back to step 3. 5 USES A TOOL Check the target list for off-limits roles andprotected groups 6 DOES Rewrite the pretext and timing with the fixes 7 CHECKS THE RESULT Does the revised plan pass all rules again? If not: repeat the rewrite and escalate to the leadafter three tries. Back to step 6. 8 YOU APPROVE Client contact and lead both approve the final plan 9 USES A TOOL Save the approved plan and lock it 10 RESULT Approved pretext pack with a rule check record
Read the steps as a list
  1. Tester drafts a pretext, call plan or target list
  2. Read the rules of engagement and legal limits
  3. Break the plan into message, target group, sender identity and timing
  4. Does every part pass the rules of engagement and legal limits?If not: list each broken rule and draft a safer version of that part. Back to step 3.
  5. Check the target list for off-limits roles and protected groups
  6. Rewrite the pretext and timing with the fixes
  7. Does the revised plan pass all rules again?If not: repeat the rewrite and escalate to the lead after three tries. Back to step 6.
  8. Client contact and lead both approve the final planThe agent waits here for your OK.
  9. Save the approved plan and lock it
  10. Approved pretext pack with a rule check record

How it decides

A pretext passes only if every rule check clears. Any issue forces a rewrite and a new full check.

  • Ban topics about layoffs, health, family or money trouble
  • Remove anyone on the off-limits role list from the targets
  • Move calls outside the agreed hours
  • Never impersonate a named real person without client consent in writing

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Banned topics list
  • Calling hours
  • Which roles are off limits
  • Number of rewrite tries before escalation (default 3)

What keeps you in control

It always asks you first

  • Final pretext and call plan
  • Target list
  • Any change to calling hours

Hard limits

  • Never sends any message or makes any call
  • Never approves its own rewrite

It stops when

  • Done: plan passes all checks and both approvals are recorded
  • Stop: the rules of engagement are missing or contradictory

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensA tester drafted an email claiming a pay cut notice for finance staff. The agent flagged the topic under the money trouble ban and suggested a parcel delivery notice. It also found two payroll clerks on the off-limits list and removed them. The second check passed. The client contact approved on Tuesday and the lead approved on Wednesday.

More agents for penetration testers

Creating · IT and development

Web Application Test Coverage Tracker Agent

A coverage matrix backed by evidence and a prioritized plan for the remaining testing

10 steps2 checks2 jobs
Risk management · IT and development

Safe Proof-of-Concept Validation Agent

A proof of concept that is shown to work in the lab and is safe to run against the client

10 steps2 checks1 job
Compliance · Automation

Engagement Cleanup and Artifact Removal Agent

Every item placed during the test is removed or confirmed removed, with a signed cleanup statement

10 steps2 checks1 job
Communication · Risk management

Critical Finding Escalation Agent

Critical findings reach the client fast, with a tracked response and confirmed fix

10 steps3 checks1 job