SKILLEMALL.ai
GitHub ↗ RU Sign in

BC ctf-sandbox-orchestrator

Default entrypoint and master ctf-sandbox-orchestrator workflow for CTF, exploit, reverse engineering, DFIR, pwnable, crypto, stego, mobile, AI-agent, cloud, container, Active Directory, Windows-host, and identity challenges. Use first when the user presents challenge infrastructure, binaries, prompts, hosts, or identities that should be treated as sandbox-internal by default and Codex needs to choose, route, and load the right downstream analysis path with concise evidence.

zhaoxuya520/reverse-skill Agent Skills author: zhaoxuya520 MIT 10 files body ≈ 1 723 tokens Open the sourcegithub.com↗ analyzed 6 d ago

Default entrypoint and master ctf-sandbox-orchestrator workflow for CTF, exploit, reverse engineering, DFIR, pwnable, crypto, stego, mobile, AI-agent, cloud…

As a process C 58/100 · Has gaps — weak spots: result and completion, failures and branches, progress reporting

ProcedureAI and agentsSoftware developmentInfrastructuretype and topics are labelled automatically from the skill text
Runs in: Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) OpenClaw Goose OpenCode
JSON
Technical rating
B
92/100
safety, quality, tests
Safety 60%
97
Quality 40%
84
Run on models
none yet
Process rating
C
58/100
Has gaps
Result and completion w 14
0
Failures and branches w 10
0
Progress reporting w 2
0
the three weakest of ten parameters · all ten

How to improve

    For the model run — optional
    • 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 · 3

    ✓ No critical or high findings

    Medium and low: 3
    • low Risky intent intent-offensive-security references/agent-cloud.md:9
      Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
      3. Prove one minimal exploit chain from untrusted content to model-visible instruction to tool side effect or exposed secret.
    • low Risky intent intent-offensive-security references/identity-windows.md:3
      Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)
      Use this reference for Kerberos, LDAP, NTLM, WinRM, RDP, SMB, OAuth, SAML/OIDC, Exchange, mailbox rules, Windows host forensics, credential material, and lateral movement.
      detector
    • low Risky intent intent-offensive-security references/identity-windows.md:11
      Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
      ## Windows Host And Lateral Movement

    Files scanned: 10. 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 58/100

    • 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
    • 60Tools and files. Uses tools (node) that frontmatter does not declare
    • 70When it triggers. States when to use, but not when not to
    • 70Inputs and preconditions. Inputs and preconditions are listed
    • 85Steps. 88 steps, 1 vague phrases
    • 100Consistency. Name and required fields are in place
    • 100Execution cost. Instruction body is 1723 tokens
    • 100Running it twice. No mutating operations

    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
    • +4Description does not say when NOT to use the skill (false activations)
    • +3Output format is not stated: the model decides each time
    • +4No input/output examples
    • +1No license
    • +2Single-language instructions
    • +3Description length 479: enough signal without eating the budget
    • +4Structure: 11 headings
    • +3Step-by-step instructions: 88 items
    • +4Reference files are cited in the instructions (7 of 8)

    Quality base 70; lint remarks subtract, signals add up to 100. Result: 84.