suggest-tasks
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes multiple local shell scripts situated in
~/.claude/hooks/analyzers/to detect project types and perform analysis (e.g.,detect-project.sh,analyze-git.sh,analyze-todos.sh). - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from project files and incorporates it into task subjects and descriptions, creating a potential path for malicious instructions to influence the agent.
- Ingestion points: Results from analyzer scripts that scan the local project directory (defined in Step 3 of
SKILL.md). - Boundary markers: Absent; findings from analyzers are parsed and used directly without delimiters or instruction-override warnings.
- Capability inventory: Shell script execution and task management through
TaskCreate. - Sanitization: No explicit validation, escaping, or filtering of the content extracted from project files is performed.
Audit Metadata