The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of scraping and processing content from public forums like Reddit and Hacker News.
Ingestion points: Data is ingested through WebSearch and WebFetch tools in 'Step 2: Scrape and Collect' from attacker-controllable sources (forum threads).
Boundary markers: Absent. The prompt for WebFetch ('Extract all complaints, feature requests...') does not include delimiters or instructions to ignore embedded commands within the fetched content.
Capability inventory: The skill has access to Bash, Write, and MemoryRecord tools, as defined in the frontmatter.
Sanitization: There is no evidence of sanitization or validation of the fetched strings before they are processed by the agent or used in scoring calculations.
[COMMAND_EXECUTION]: The 'Memory Protocol' section instructs the agent to execute a local JavaScript file using Node.js via the command line.
Evidence: node .claude/lib/memory/memory-search.cjs "forum monitor pain points trends".
Context: While this appears to be a local utility for memory management, it demonstrates the use of shell execution to run scripts.
[COMMAND_EXECUTION]: The skill utilizes scheduled task creation for persistence across sessions.
Evidence: The 'Cron Integration' section uses CronCreate to schedule the execution of the forum monitor scan every Monday.
Context: This is the primary intended purpose of the skill for periodic research, but it represents a persistence mechanism.

forum-monitor