The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and process data from external web sources, which presents a surface for indirect prompt injection if the ingested content contains malicious instructions aimed at the AI.
Ingestion points: Fetches content using belt app run tavily/search-assistant and belt app run exa/search (SKILL.md).
Boundary markers: The skill's markdown templates do not explicitly provide delimiters or instructions for the agent to ignore instructions embedded within the curated links or summaries.
Capability inventory: The skill has access to the belt CLI tool, which can run various apps including search assistants and social media posters (x/post-create).
Sanitization: No specific sanitization or filtering mechanisms for the content retrieved from search results are defined in the instructions.
[EXTERNAL_DOWNLOADS]: The skill utilizes and recommends the installation of external tools and dependencies associated with the platform ecosystem.
Recommends installing the belt CLI via npx skills add belt-sh/cli.
References external installation documentation hosted on GitHub at raw.githubusercontent.com/inference-sh/skills/.
Suggests adding related skills from the inference-sh/skills repository.
[COMMAND_EXECUTION]: The skill provides various shell commands for the agent to execute using the belt utility.
Includes commands for authentication (belt login), content discovery, and automated social media posting.

newsletter-curation