The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several internal shell scripts (generate-slides.sh, publish-carousel.sh, check-analytics.sh) and CLI tools including playwright and uv to execute its workflow.
[COMMAND_EXECUTION]: The skill implements a persistence mechanism by instructing the agent to 'Self-Schedule' its own recurring execution using system task schedulers like cron, allowing it to maintain a daily publishing cycle without human oversight.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it scrapes arbitrary website URLs using Playwright and uses the extracted content to drive its autonomous decision-making.
Ingestion points: Untrusted data is ingested from external URLs via analyze-web.js and Playwright.
Boundary markers: There are no instructions for the agent to use delimiters or ignore embedded commands within the scraped website data.
Capability inventory: The agent has capabilities to execute local scripts, write files (learnings.json), and perform network requests to external APIs.
Sanitization: The skill lacks any logic for escaping or validating the content extracted from external websites before it influences the agent's content generation pipeline.
[EXTERNAL_DOWNLOADS]: The skill triggers the download and installation of the Chromium browser binary through the playwright install chromium command.
[DATA_EXFILTRATION]: The skill performs network operations to api.upload-post.com, a non-whitelisted third-party domain, to transmit generated content and retrieve analytics data.

marketing-carousel-growth-engine