pp-botsee
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install an external command-line interface (CLI) from public package registries.
- Evidence: Installation via NPM using
npx -y @mvanhorn/printing-press-library install botsee --cli-only. - Evidence: Alternative installation via Go using
go install github.com/mvanhorn/printing-press-library/library/marketing/botsee/cmd/botsee-pp-cli@latest. - [COMMAND_EXECUTION]: The skill drives a local binary (
botsee-pp-cli) to perform its primary tasks. - Evidence: Commands such as
ai-visibility-audit,site-config, andrecommendationsexecute the binary to interact with the BotSee API and local data stores. - [DATA_EXFILTRATION]: The CLI tool includes a feature to deliver output to arbitrary external URLs, which could be used to transmit data.
- Evidence: The
--deliver webhook:<url>flag allows the tool to POST its output to a user-provided webhook endpoint. - [PROMPT_INJECTION]: The skill analyzes data from external domains, creating a potential surface for indirect prompt injection.
- Ingestion points: Untrusted data enters the agent context when
ai-visibility-audit <url>fetches and analyzes content from the provided URL (SKILL.md). - Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the fetched content are defined in the instructions.
- Capability inventory: The agent has the ability to execute shell commands (
botsee-pp-cli), write files (--output), and perform network operations (--deliver webhook) (SKILL.md). - Sanitization: There is no mention of sanitization or filtering of the data retrieved from external websites before processing.
Audit Metadata