More screenshots
Works with the AI Agents you already use
Pentest Tools: Authorized Security Testing Toolkit
Plan and run authorized penetration tests that hold up: a hard authorization gate, phased recon-to-credential workflow, and evidence logging that makes every finding reproducible.
$5
Pentest Tools: Authorized Security Testing Toolkit
Example session with this skill installed
We have written authorization to test our staging environment, scan-only, this week. Plan the engagement and show me what the output looks like.
- Read your context and instructions
- Compiled the pentest tools
- Generated the document
ENGAGEMENT PLAN + FIRST FINDING — produced by applying the skill.
Gate check: written scope on file (staging hosts only, scan-only techniques, business-hours window, named contact). Blast radius: the staging domain resolves to a dedicated staging cluster, no shared tenants; two third-party services the app calls are noted as observe-only. Evidence log opened; scope document is entry one. Pre-flight checklist 8/8.
Phase 1 — Recon. Subdomain enumeration returns 14 names; HTTP probing finds 6 live. Deep service scans on the 6 show five current services and one host, admin staging, running an admin panel two major versions behind.
Phase 2 — Vulnerability ID. Nuclei high-severity templates flag a known CVE against that panel. Manual verification: the banner and the version endpoint both confirm the vulnerable release, not a backported patch — scanner-only status removed, finding admitted.
Finding F-01 (rubric 10/10). Command: nuclei high-severity template set against the admin staging host, followed by manual version checks. Verification: version endpoint response reproduced twice, one hour apart. Impact, one sentence: unauthenticated access to this panel would expose the customer records mirrored into staging. Recommendation: upgrade the panel or take it off the network until patched; retest after the fix.
Phases deliberately not run: web application testing beyond identification, and all credential testing — both outside the scan-only scope. Listed in the report as untested surface, not implied coverage. If the owner wants them, that is a written scope amendment, scheduled inside a new window.
pentest-tools-authorized-security-testin.pdf
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
A penetration test is only as good as its paperwork and its evidence. This skill runs authorized engagements end to end, starting where most testing goes wrong: a hard authorization gate that requires written scope naming the targets, the allowed techniques, and the time window before any packet is sent — plus a blast-radius check for the shared hosts and third-party services that scopes forget. The work then follows a four-phase arc that never skips ahead: passive-then-active reconnaissance that produces a signed-off asset inventory; vulnerability identification with template scanning (Nuclei), second-opinion misconfiguration scans, and version-to-CVE mapping, with a triage rule that no scanner hit becomes a finding until a human verifies it; web application testing covering injection, content discovery, XSS, and the auth and session flows scanners are weak at; and credential and hash testing that stays strictly inside what the scope authorizes, with lockout thresholds respected and owners warned. Three tooling routes (native CLI, a Dockerized MCP bundle, per-tool MCP servers) let you match the setup to the engagement, and an evidence discipline runs through everything: every command and output lands in a timestamped log, and every finding carries the exact command, the raw output, the manual verification step, and the business impact in one sentence. You get a fill-in engagement plan template, an eight-item pre-flight checklist, a finding quality rubric (ship at 8+/10, no zero on verification), edge-case handling for out-of-scope discoveries and fragile targets, and a complete worked example of a scan-only staging engagement — including the phases it deliberately did not run.
- It cannot create authorization: without a written scope naming targets, techniques, and a window, the correct output of this skill is a stopped engagement.
- It covers identification and verification, not exploit development or post-exploitation, which are separate disciplines with separate gates.
- Scanner and CVE data go stale; version-to-CVE mapping is only as current as the feeds you check it against on the day.
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 24 days ago