The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions direct the agent to execute a Python script using a bash command that interpolates a [brand_name] variable. If this variable is extracted from untrusted sources (like the markdown reports it parses) and contains shell metacharacters (e.g., ;, &&, |), it could lead to arbitrary command execution.
[EXTERNAL_DOWNLOADS]: The skill requires the reportlab Python package to be installed via pip. While reportlab is a standard library for PDF creation, it represents an external dependency required for the skill to function.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it parses untrusted data from local markdown files (GEO-*.md) to populate the report's JSON structure and command-line arguments.
Ingestion points: Local markdown files matching the pattern GEO-*.md and potentially user-provided URLs.
Boundary markers: Absent. The skill does not instruct the agent to use delimiters or ignore embedded instructions when parsing these files.
Capability inventory: The skill uses the Bash tool to execute scripts and the Write tool to create temporary files, providing a path for malicious instructions to reach the system shell.
Sanitization: None. There is no requirement or logic provided to escape or validate the content extracted from the markdown reports before it is used in the command execution step.

geo-report-pdf