programmatic-seo
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Python scripts to perform SEO operations.
- Evidence:
python scripts/dfs_client.pyandpython scripts/report_publish.pyare used to interface with SEO APIs and generate formatted deliverables. - The skill passes user-supplied inputs such as
<template head term>and<user-domain>as command-line arguments. This is standard for the skill's utility but requires the execution environment to handle potential shell metacharacters in user input. - [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it is designed to ingest and process external datasets or data sources provided by the user.
- Ingestion points: The skill requires a "dataset or data source" to generate pages at scale.
- Boundary markers: No explicit prompt delimiters are defined for the dataset content in the instructions.
- Capability inventory: The skill has the capability to write files (
PSEO-PLAN-*.md) and execute Python scripts (dfs_client.py,report_publish.py). - Sanitization: The skill includes a "Gate the dataset" process (Step 2) that performs pass/fail tests on data uniqueness and depth before template design, which acts as a logical validation layer.
Audit Metadata