The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest and process data from external search providers (Tavily and Exa) to assist with research and fact-checking. This creates a surface for indirect prompt injection where malicious content from the web could potentially influence the agent's writing or behavior.
Ingestion points: SKILL.md (via infsh app run tool calls for research and market data).
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded prompts within the retrieved search data.
Capability inventory: Access to Bash(infsh *) for executing search queries and platform commands.
Sanitization: No explicit sanitization or validation of the external content is described before it is used in the writing process.
[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the inference.sh CLI and additional extensions using npx skills add inference-sh/skills@agent-tools. These resources are part of the vendor's infrastructure (tul-sh) and are used to provide the research and fact-checking capabilities described in the documentation.
[COMMAND_EXECUTION]: The skill requires access to the Bash tool to execute infsh commands. These commands are used for session management (infsh login) and running research applications (infsh app run) to support the press release writing workflow.

press-release-writing