The Agent Skills Directory

[PROMPT_INJECTION]: The skill acts as an intermediary for processing arbitrary user-provided Markdown content to generate HTML and send data to external webhooks. This establishes an attack surface for indirect prompt injection where malicious instructions embedded in user content could potentially influence the agent's actions during the formatting or delivery process.
Ingestion points: User-supplied Markdown text defined in usage examples within SKILL.md.
Boundary markers: No explicit delimiters or warnings are specified to separate user-provided content from the skill's operational instructions.
Capability inventory: Includes file system writes for HTML previews (/tmp/slack-formatter), system clipboard access for pasting, and network operations for Slack webhook delivery.
Sanitization: The skill does not define specific sanitization or escaping protocols for the content conversion process, relying instead on the agent's internal translation logic.
[COMMAND_EXECUTION]: The skill directs the agent to execute system-level actions such as writing HTML files to the temporary directory and using clipboard utilities. These commands are integral to the core functionality of providing Slack-themed previews and easy pasting.
[DATA_EXFILTRATION]: The skill supports sending content to an external Slack webhook defined by the CCH_SLA_WEBHOOK environment variable. While this involves sending data to an external endpoint, it targets a well-known service and is a documented feature of the skill's communication capabilities.

slack-message-formatter