More screenshots
Works with the AI Agents you already use
CLI Binary Triage with radare2
Dissect PE, ELF, Mach-O, DEX, or WASM files from the terminal — file identity, imports, functions, cross-references — and verify behavior before concluding anything.
$5
CLI Binary Triage with radare2
Example session with this skill installed
A suspicious updater binary arrived with no documentation. Triage it from the terminal: what is it, what does it call, and is the update mechanism safe to trust?
- Read your context and instructions
- Compiled the cli binary triage
Identity first, no session needed: a 64-bit ELF, dynamically linked, stripped, with imports dominated by networking and process-spawning calls — unusual weight for an updater. Orientation by strings surfaced a hardcoded update URL, a version-check path, and a fallback address that did not match the vendor's domain. Function navigation followed the cross-references from the network calls to a download routine with no signature check: the file is fetched, written to disk, and executed without verifying anything about it. The companion tools confirmed the picture: the binary's embedded certificate list was empty, and the download path had no integrity check at any stage. Verdict: do not trust the update mechanism. The binary is an updater in shape only — its download-and-execute path with no verification is the finding, and the report names the exact functions and addresses so the next analyst can reproduce it in minutes. No bytes were patched; the request was triage, and triage is what was delivered. The report was filed with the binary's hash, so a second analyst can confirm the same findings on the same file — and the verdict stands on the evidence in the report, not on the analyst who wrote it.
Connects securely to your tools. The creator never sees your data.
About this skill
A binary you cannot open in a GUI still has to be understood, and the terminal is where that understanding starts. This skill drives the radare2 toolchain for CLI-first binary triage: identifying a file and its imports in two minutes without opening a session, orienting by strings, navigating functions and cross-references, and patching bytes only when the request explicitly says so. The discipline is evidence-first: every conclusion traces to something observed in the binary, behavior is verified before it is claimed, and the report separates what was seen from what was inferred. It covers the companion tools that fill radare2's gaps and a reporting shape that makes a triage pass useful to the next person who reads it.
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.
Triage is not a full reverse-engineering engagement — it answers what the binary is and what it does at a first pass, not every behavior it can exhibit.
Stripped and packed binaries limit what static analysis can say; the skill names that limit instead of guessing past it.
Patching is covered only as a deliberate, explicitly requested step — never as part of triage.
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 3 days ago
Permissions
No special permissions declared or detected