consult-setup
Pass
Audited by Gen Agent Trust Hub on Jul 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the 'Bash' tool to run local scripts such as 'engagement-init.sh' and 'discover-projects.sh' from the plugin root. These scripts perform scaffolding and maintain a global registry at '$HOME/.claude/cogni-consult-projects.json'.
- [PROMPT_INJECTION]: The skill defines a workflow that ingests untrusted user input and interpolates it into shell commands, creating an indirect prompt injection surface.
- Ingestion points: Engagement name, client, and outcome provided by the user in Step 1.
- Boundary markers: Absent. The skill does not instruct the agent to escape or sanitize these inputs before interpolation into shell commands.
- Capability inventory: The skill uses 'Bash', 'Write', 'Edit', and 'Skill' tools across its workflow.
- Sanitization: The skill recommends deriving a kebab-case slug for one argument but passes the raw engagement name as a second argument to the init script.
Audit Metadata