The Agent Skills Directory

[NO_CODE]: The analyzed skill consists entirely of markdown documentation, logic descriptions, and JSON/markdown file templates. It does not include any Python, Node.js, or shell scripts.
[SAFE]: The external links provided (The Coaching Tools Company and GoalsWon) are for educational reference and best practices. These are well-known resources in the coaching industry and do not present a security risk.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it processes untrusted user input to generate challenge configurations and schedules.
Ingestion points: User responses during the 5-phase onboarding flow (e.g., challenge names, milestones, punishment descriptions).
Boundary markers: Absent; user input is directly interpolated into file templates without explicit delimiters or instructions to the agent to ignore embedded commands.
Capability inventory: Writing configuration files to the local file system at ~/.openanalyst/challenges/.
Sanitization: Not explicitly defined in the logic; the skill relies on the underlying agent's standard safety filters.
[DATA_EXPOSURE]: The skill specifies the creation of local files in the user's home directory (~/.openanalyst/). These files store user-defined goals and commitments. This is considered standard functional behavior for an accountability application and does not involve external exfiltration.

challenge-onboarding