The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes raw, untrusted data from social media posts.
Ingestion points: The skill scrapes tweet content from the X platform's 'For You' feed using browser automation as described in references/scraping-guide.md.
Boundary markers: The template in assets/digest-template.md uses markdown blockquotes (> {{content}}) for tweet content, which does not provide a robust security boundary to prevent the model from following instructions embedded in the text.
Capability inventory: The skill has the ability to write files to the local system and execute browser automation commands.
Sanitization: No sanitization or filtering logic is present to identify or remove adversarial instructions from the scraped tweets before they are passed to the agent.
[COMMAND_EXECUTION]: The skill requires the agent to execute various commands for its workflow.
The instructions in SKILL.md and references/share-card-guide.md involve running Playwright scripts and browser extension commands (npm run start-extension).
[EXTERNAL_DOWNLOADS]: The skill depends on external tools and specific browser environments.
It requires the dev-browser skill and playwright to be installed and running in the execution environment.

x-ai-digest