The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Node.js script for material collection using the shell.
Evidence: node "$SKILLS_ROOT/content-planner/scripts/wechat_search.js" "topic keywords" -n 10 in SKILL.md.
Risk: The "topic keywords" are derived directly from user input or external files (content_calendar.json). If these inputs contain shell metacharacters (e.g., ;, &, |), it could lead to arbitrary command execution on the host system.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection via the data it processes.
Ingestion points: The skill reads topic descriptions and metadata from content_calendar.json and direct user input (SKILL.md).
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when interpolating these inputs into the workflow.
Capability inventory: The skill has the ability to execute shell commands (via node) and write files to the drafts/ directory.
Sanitization: There is no evidence of input validation or escaping for the data used in command arguments or content generation.

article-writer