CB reverse-engineering
Provides reverse engineering techniques. Use when the main job is to understand how a compiled, obfuscated, packed, or virtualized target works before exploiting or solving it, including binaries, APKs, WASM, firmware, custom VMs, bytecode, malware-like loaders, and anti-debug or anti-analysis logic. Do not use it when the vulnerability is already understood and the remaining task is exploitation; use pwn instead. Do not use it for pure web workflows, log or disk forensics, or standalone crypto problems unless reversing the implementation is the real blocker.
Provides reverse engineering techniques.
As a process B 68/100 · Nearly there — weak spots: result and completion, failures and branches, running it twice
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill asks for more than the task needs: broad tool access, credential environment variables, binaries. Every extra permission widens the damage from a mistake or a compromise.
Narrow allowed-tools and the variable list to the minimum; replace binaries with readable sources or scripts.
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 16
✓ No critical or high findings
Medium and low: 16
-
medium Dangerous commands
cmd-pipe-to-shellreferences/nonpe-format-cookbook.md:90Downloads and executes remote code from an unrecognised host (pipe to shell) (documentation of a security skill)| **AS** | 循环 curl|sh / 远程拉脚本、非常规 C2 URL | 提 URL；分析下载体是否含格机命令；记临时路径 | E-android-wiper-backdoor | P0 |
security skill -
medium Broad scope
meta-broad-allowed-toolsSKILL.md:1Broad tool permissions pre-approved: Bashallowed-tools: Bash Read Write Edit Glob Grep Task WebFetch WebSearch
-
low Risky intent
intent-offensive-securityfield-notes.md:52Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- [Hash-Resolved Imports / No-Import Ransomware](#hash-resolved-imports--no-import-ransomware)
-
low Risky intent
intent-offensive-securityfield-notes.md:259Offensive-security / dual-use content (legitimate for authorised testing; review intended use)### Hash-Resolved Imports / No-Import Ransomware
-
low Risky intent
intent-offensive-securityfield-notes.md:260Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)**Pattern:** Binary with zero visible imports resolves APIs via symbol name hashing at runtime. Skip the hash reversing — hook OpenSSL functions via `LD_PRELOAD` in Docker to capture AES keys directly
detector -
low Risky intent
intent-offensive-securitylanguages-compiled.md:200Offensive-security / dual-use content (legitimate for authorised testing; review intended use)**Pattern (see-two):** A Go-compiled C2 client has a UUID embedded via `-ldflags -X`. The C2 server uses mTLS for authentication. To enumerate other clients and their files, patch the UUID to register
-
low Risky intent
intent-offensive-securitylanguages-compiled.md:205Offensive-security / dual-use content (legitimate for authorised testing; review intended use)3. Register with the C2 server using the patched binary (mTLS certs are embedded or in distfiles)
-
low Secrets in code
secret-high-entropy-tokenlanguages-compiled.md:343High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)swift demangle 's14M…_tF'
quoted -
low Risky intent
intent-offensive-securitypatterns-ctf.md:17Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- [Hash-Resolved Imports / No-Import Ransomware (BSidesSF 2026)](#hash…026)
-
low Risky intent
intent-offensive-securitypatterns-ctf.md:417Offensive-security / dual-use content (legitimate for authorised testing; review intended use)## Hash-Resolved Imports / No-Import Ransomware (BSidesSF 2026)
-
low Risky intent
intent-offensive-securitypatterns-ctf.md:432Offensive-security / dual-use content (legitimate for authorised testing; review intended use)// hook_crypto.c — captures AES key used by the ransomware
-
low Risky intent
intent-offensive-securitypatterns-ctf.md:457Offensive-security / dual-use content (legitimate for authorised testing; review intended use)# Run in Docker container (ransomware may be destructive!)
-
low Risky intent
intent-offensive-securitypatterns-ctf.md:459Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (quoted — discussed, not commanded)bash -c "LD_PRELOAD=./hook.so ./ransomware; xxd /tmp/aes_key.bin"
quoted -
low Secrets in code
secret-high-entropy-tokenplatforms.md:113High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)swift demangle 's14M…_tF'
quoted -
low Risky intent
intent-offensive-securitySKILL.md:65Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- [patterns-ctf.md](patterns-ctf.md) - Competition-specific patterns (Part 1): hidden emulator opcodes, LD_PRELOAD key extraction, SPN static extraction, image XOR smoothness, byte-at-a-time cipher, m
A further 1 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 24. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 68/100
- 0Result and completion. Does not say what the result is
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 1 mutating operations with no state check
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Tools and files. Tools declared in frontmatter
- 100Steps. 45 steps
- 100When it triggers. States when to use and when not to
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2668 tokens
- low 14 top-level sections: this looks like several domains in one skill
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +3Output format is not stated: the model decides each time
- -44 reference files, but SKILL.md never points to them: the model will not open them
- +2Single-language instructions
- +4Description says when NOT to use the skill
- +3Description length 565: enough signal without eating the budget
- +4Structure: 15 headings
- +3Step-by-step instructions: 45 items
- +4Has examples (8 code blocks)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 85.