skills/fradser/skills/start-hotfix/Gen Agent Trust Hub

start-hotfix

Warn

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The shell script scripts/start-branch.sh contains a command injection vulnerability. Specifically, the $VERSION variable is interpolated into a sed command without escaping the / delimiter, which allows for the execution of arbitrary sed commands, including file writes or potentially shell command execution. Additionally, the $NAME variable is used unquoted in an echo statement, which could lead to subshell expansion and command execution if the input contains backticks or $(...) syntax.\n- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection via repository metadata. As described in references/gitflow-start-pipeline.md, the agent analyzes commit history using git log to determine versioning logic. This ingestion point lacks boundary markers or instructions to ignore embedded commands, allowing an attacker to influence agent decisions through malicious commit messages. The skill's capabilities include shell script execution and remote repository modification.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 18, 2026, 03:52 PM
Security Audit — agent-trust-hub — start-hotfix