The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its content fetching capabilities.
Ingestion points: Data is ingested from external URLs via the WebFetch tool and from user-provided files.
Boundary markers: There are no explicit instructions or delimiters (such as XML blocks or 'ignore' directives) to prevent the agent from obeying instructions embedded within the fetched articles or social media posts.
Capability inventory: The skill possesses the Bash, Write, and WebFetch tools, providing a significant capability surface if an injection occurs.
Sanitization: No sanitization or validation logic is defined for the content retrieved from the web before it is processed by the analysis pipeline.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute several local Node.js scripts for analysis and memory management.
Evidence: Commands include node .claude/tools/cli/post-analyzer.cjs and node .claude/lib/memory/memory-search.cjs.
Context: While these scripts are part of the skill's defined infrastructure, they represent the execution of local logic that is not visible in the provided markdown file.

content-analyzer