paw-mkt-seo
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted data from external websites during SERP and competitor analysis sessions.
- Ingestion points: Content is retrieved via 'agent-browser get text body' in 'references/capability-research.md'.
- Boundary markers: No specific instructions are provided to the agent to ignore or delimit embedded commands in the fetched content.
- Capability inventory: The agent utilizes 'agent-browser' for web interaction and can write SEO deliverables to the local file system.
- Sanitization: The skill does not implement validation or filtering of the ingested external text.
- [COMMAND_EXECUTION]: The skill uses shell commands and scripts to prepare the environment and discover browser configurations.
- Evidence: Instructions in 'references/shared-patterns.md' and 'references/capability-research.md' involve running local scripts and installing tools via npm.
- [EXTERNAL_DOWNLOADS]: The skill downloads and installs necessary browser automation tools from public repositories.
- Evidence: 'references/shared-patterns.md' describes the installation of 'agent-browser' from a well-known service's repository.
Audit Metadata