skills/karlorz/agent-skills/simplify/Gen Agent Trust Hub

simplify

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks because it ingests and processes untrusted source code through git diffs. A malicious actor could embed instructions within code comments or logic to influence the agent's behavior.
  • Ingestion points: The skill uses git diff to ingest code changes for review in SKILL.md.
  • Boundary markers: No explicit delimiters or instructions are used to distinguish code data from the agent's operational instructions.
  • Capability inventory: The skill has the capability to write files ("Fix worthwhile issues directly") and execute arbitrary validation commands ("tests, lint, typecheck, or a targeted build step") as described in SKILL.md.
  • Sanitization: There is no evidence of sanitization or filtering of the ingested code content.
  • [COMMAND_EXECUTION]: The skill instructions direct the agent to execute shell commands for both scoping changes (git diff) and validating results (tests, lint, build). If the skill is used on a repository with malicious build configurations or test scripts, these commands could lead to local code execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 12:49 PM