The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requires the agent to execute a shell command using curl to send a notification to a local service (http://localhost:31337/notify) upon invocation.
[COMMAND_EXECUTION]: The skill instructs the agent to append execution logs to a specific file on the local file system (~/.claude/LIFEOS/MEMORY/SKILLS/execution.jsonl) using shell redirection.
[PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection by design, as its primary function is to ingest and process arbitrary user-provided content such as story notes, character sketches, and draft chapters.
Ingestion points: The skill reads folders of notes and draft scenes from the user's local environment to build the 'Story Bible' and 'spine'.
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from being influenced by instructions potentially embedded within the ingested story notes.
Capability inventory: The agent has the capability to execute shell commands (curl, echo) and invoke other skills (ISA, BeCreative, Task).
Sanitization: The skill lacks explicit sanitization or validation logic for the external text content it processes.

WriteStory