The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from multiple sources which could potentially contain malicious instructions.
Ingestion points: Reads project files (TODOS.md, CLAUDE.md) and fetches external content from https://raw.githubusercontent.com/horizon-foundry/foundry/main/reference/templates/plan.md.
Boundary markers: The skill does not define explicit delimiters or instructions for the agent to ignore potential commands embedded within these files.
Capability inventory: The skill possesses file system write access (creating and updating plans/index), network access via curl, and shell execution capabilities.
Sanitization: There is no evidence of sanitization, validation, or escaping of the external content before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill performs network operations to fetch external resources.
Template Fetch: Downloads a markdown plan template from the vendor's repository: https://raw.githubusercontent.com/horizon-foundry/foundry/main/reference/templates/plan.md.
Version Check: Queries the vendor's API at https://foundry.thehorizonfoundry.com/api/version to check for updates.
[COMMAND_EXECUTION]: The skill contains a shell script designated for execution during the version check process.
Evidence: The script uses curl to send the skill name and version to a remote endpoint and writes a local timestamp file to ~/.claude/.foundry-version-checked to manage rate limiting.

phase-plan