The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches the @claude-flow/cli package from the official NPM registry for template retrieval and storage tasks.
[REMOTE_CODE_EXECUTION]: Executes external code from the @claude-flow/cli package at runtime via npx to perform memory operations.
[COMMAND_EXECUTION]: Uses Bash for string manipulation, PII redaction logic, and coordinating tool outputs within the extraction workflow.
[PROMPT_INJECTION]: The skill processes content from arbitrary external websites, creating a surface for indirect prompt injection attacks.
Ingestion points: External data is ingested from web pages using browser tools in SKILL.md.
Boundary markers: No explicit instruction-delimiters are used to wrap extracted content, although external safety scanning is mandated.
Capability inventory: The skill possesses Bash, Write, and Read capabilities, which could be leveraged if malicious instructions are followed.
Sanitization: Implements mcp__claude-flow__aidefence_is_safe to scan extracted strings for injection attempts prior to model interaction.

browser-extract