The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes multiple local TypeScript tools and shell commands using the bun runtime (e.g., InterviewScan.ts, TelosFreshness.ts, GenerateTelosSummary.ts). These commands are used for routing logic and content management within a hidden directory structure (~/.claude/LIFEOS/TOOLS/). It also performs network requests to a local notification service (localhost:31337) using curl.
[PROMPT_INJECTION]: The skill demonstrates a clear surface for indirect prompt injection by ingesting and acting upon the content of various user-controlled files.
Ingestion points: The workflow reads seven distinct 'constitutional' files including TELOS.md, PROJECTS.md, and DA_IDENTITY.md, as well as their metadata.
Boundary markers: Absent; instructions direct the agent to incorporate the raw content of these files directly into conversational loops without clear delimiters or warnings to ignore embedded instructions.
Capability inventory: The agent has the capability to write to these files and the environment configuration (.env) via the Edit tool, and can execute shell commands via bun.
Sanitization: There is no evidence of sanitization or structural validation for the content read from these files before it is processed by the agent's logic.

Interview