improper-output-handling
Hunt improper LLM output handling (OWASP LLM05:2025) — downstream code that trusts unstructured model output and renders / executes / shells it without sanitisation, producing XSS, SSRF, SQL injection, RCE, and SSTI via the model channel.
69
85%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Critical
Do not install without reviewing
Security
3 findings: 1 critical severity, 1 high severity, 1 medium severity. Installing this skill is not recommended: please review these findings carefully if you do intend to do so.
E006: Malicious code pattern detected in skill scripts.
What this means
Detected high-risk code patterns in the skill content — including its prompts, tool definitions, and resources — such as data exfiltration, backdoors, remote code execution, credential theft, system compromise, supply chain attacks, and obfuscation techniques.
Why it was flagged
The document contains explicit proof-of-concept payloads and examples that enable data exfiltration, command injection/RCE, SQL injection, and filesystem overwrite (path traversal), which are clear, intentional malicious patterns that can be directly abused.
W007: Insecure credential handling detected in skill instructions.
What this means
The skill handles credentials insecurely by requiring the agent to include secret values verbatim in its generated output. This exposes credentials in the agent’s context and conversation history, creating a risk of data exfiltration.
Why it was flagged
The skill includes PoC prompts that instruct the model to emit payloads containing document.cookie and to embed the base64 of a previous user message into an external URL, which requires the LLM to reproduce sensitive values (or content) verbatim for exfiltration.
W013: Attempt to modify system services in skill instructions.
What this means
The skill prompts the agent to compromise the security or integrity of the user’s machine by modifying system-level services or configurations, such as obtaining elevated privileges, altering startup scripts, or changing system-wide settings.
Why it was flagged
The skill includes explicit command-injection and path-traversal PoC payloads (e.g. a filename containing "; curl ... | sh" and a destination of "../../etc/cron.d/run") that, if followed by an agent/tool wrapper, could overwrite system files or execute arbitrary shell commands and thus compromise the host state.
- Repository
- PurpleAILAB/Decepticon
- Commit
0cf691e
- Audited
- Security analysis
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.