More screenshots
Works with the AI Agents you already use
Hardware Security Research: Interfaces and Debug Ports
Assess an authorized embedded device methodically: document the board, identify UART and debug interfaces by measurement, capture evidence, and verify firmware images by hash.
$5
Hardware Security Research: Interfaces and Debug Ports
Example session with this skill installed
Assess a battery-powered environmental sensor you own and are authorized to open. Questions: does it expose a debug console, is production JTAG available, and can a firmware image be captured reliably? Destructive work is not authorized. The board has an unpopulated four-pin header and an eight-pin flash package.
- Read your context and instructions
- Compiled the hardware security research
Intake: device photographed before opening; authorization limits recorded as enclosure opening, power-on observation, receive-only console capture, interactive access only if voltage and pinout are confirmed, debug identification allowed, chip removal prohibited. Battery state and board revision entered in the evidence log.
Board survey: main processor, 8 MB SPI flash, power-management device, and radio module identified from markings. The four-pin header sits near the processor. Ground was confirmed by continuity to the shield. One pin measured 3.3 V; one remained near ground; one idled high and produced bursts during reset, supporting a probable transmit identification. No pin was connected until those measurements were recorded.
Read-only capture: adapter ground connected and adapter receive connected to the probable transmit pin, with the adapter transmit left disconnected. At 115200 baud, the boot log showed bootloader and application versions, partition names, and a message that the interactive console was disabled in the production build. No credentials or key material appeared in the captured log. Result: UART is exposed physically, but the supplied evidence does not support an unauthenticated shell finding.
Debug route: test points and processor documentation supported an SWD identification attempt. The chain returned a valid processor identifier, but halt and memory access were not available in the tested state. Finding recorded narrowly as “SWD responds to identification; access appears locked; memory contents not assessed.” No unlock or bypass technique was attempted because it was outside the stated method and unnecessary for the question.
Firmware route: in-circuit reading was judged possible but not needed to answer the first two questions, and clip pressure risked disturbing a battery-backed board. It was deferred rather than escalated. Close-out lists the pin map with confidence levels, raw log settings, photographs, locked-interface result, and the device restored with its enclosure and battery reconnected.
Connects securely to your tools. The creator never sees your data.
About this skill
Opening a device and connecting to the first promising row of pins is how boards get damaged and findings become anecdotes. This skill provides an authorization-first method for embedded hardware assessment. It starts with scope, hazards, power, and preservation requirements, then moves from photographs and component identification to measured ground and voltage discovery, receive-only UART capture, cautious JTAG or SWD enumeration, and least-invasive firmware extraction. Each escalation has a decision rule, so interactive access, debug probing, in-circuit reading, and chip removal are deliberate choices rather than momentum. The package includes an evidence log template, an invasiveness ladder, an interface triage checklist, and a finding-quality rubric that separates an observation from a supported security finding. It also covers locked interfaces, encrypted images, intermittent reads, boot-log secrets, restoration, and the honest reporting of negative results. Use it for a lab assessment of your own product, a returned unit, a development board, or another device whose owner has clearly authorized the work and its limits.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.
- Pin identification and safe probing depend on the actual board, power design, and tools available; no written workflow can guarantee a pinout.
- Locked, encrypted, or undocumented devices may yield limited results, which should be reported as such.
- It does not provide authorization, legal advice, or a substitute for electrical-safety training.
- Firmware analysis after extraction is outside this skill except for integrity, scope, and evidence handling.
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 buying a skill.
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
Permissions
No special permissions declared or detected