agency-dashboard
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes local Python scripts (
campaign-tracker.py,execution-tracker.py, andteam-manager.py) located within the plugin's root directory to aggregate metrics and assess team capacity.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local configuration files and user-provided outputs, which presents a surface for potential instruction injection if the data source is compromised.\n - Ingestion points: Reads brand profiles, guidelines, and manifests from the
~/.claude-marketing/brands/directory and processes user-provided CLI usage output.\n - Boundary markers: No explicit delimiters or guardrail instructions are used to isolate the ingested content from the agent's logic.\n
- Capability inventory: The skill utilizes subprocess execution and directory enumeration to perform its core functions.\n
- Sanitization: There is no evidence of sanitization or validation of the ingested JSON content or the directory names (slugs) before they are used in scripts or prompts.
Audit Metadata