morning-brief

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes dynamic context injection (!command syntax) in the SKILL.md file to verify the environment (checking for gh and jq availability) and to resolve the current repository name at load time.
  • [COMMAND_EXECUTION]: The skill executes a local Bash script (morning-brief.sh) which invokes standard system utilities including gh, jq, grep, sed, awk, and date to process and format repository data.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it reads and prints GitHub issue bodies and comments. Since these are external and potentially attacker-controlled, they could contain instructions intended to influence the agent. However, the skill is designated as read-only and does not provide tools for the agent to take action based on this data, mitigating the risk.
  • [DATA_EXFILTRATION]: The skill reads GitHub metadata (PRs, issues). This data is strictly limited to the repository context and is displayed as standard output for the user's view, which is the primary intended function of the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 07:55 PM
Security Audit — agent-trust-hub — morning-brief