The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and summarize data from local plugin files, creating a surface for indirect prompt injection if those files contain malicious instructions.
Ingestion points: Local artifact files located at plugin-data/hermes-achievements/ and plugins/hermes-achievements/.
Boundary markers: The skill includes explicit instructions to "Separate prepared guidance from observed platform, runtime, connector, file, memory, or delivery evidence" and mentions a "prepared-vs-observed boundary."
Capability inventory: The primary capability is observability (summarization and reporting). It does not appear to use high-risk capabilities like arbitrary command execution or network exfiltration.
Sanitization: The skill does not provide specific instructions for sanitizing or escaping the data read from these local files before it is summarized for the user.

omh-achievements