csharp-diagnose

Warn

Audited by Gen Agent Trust Hub on Apr 9, 2026

Risk Level: MEDIUMCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill constructs and executes shell commands that include variables such as <solution>.sln and <paths> derived from the file system. If an attacker can control file or directory names within the repository (e.g., via a malicious Pull Request), they could inject shell metacharacters to execute arbitrary commands.
  • Evidence in SKILL.md:
  • jb inspectcode <solution>.sln ... (Step 3)
  • similarity-csharp -p <paths> ... (Step 3)
  • jq ... results.sarif (Step 4)
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from tool outputs and source code files to generate a report, which could be used to influence the agent's subsequent behavior if the code contains malicious instructions.
  • Ingestion points: Reads results.sarif and console output from similarity-csharp (Step 4).
  • Boundary markers: Absent; the skill does not use delimiters or instructions to ignore embedded content in the analyzed data.
  • Capability inventory: Shell execution (jb, similarity-csharp, jq, rm) and file system access.
  • Sanitization: None detected for the data processed from the SARIF results or console output.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 9, 2026, 05:40 PM
Security Audit — agent-trust-hub — csharp-diagnose