The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external content which may include untrusted data such as practice materials and user-generated activity summaries.
Ingestion points: The tools ielts_review_read_snapshot (for reading material snapshots) and ielts_footprints_list (for reading activity descriptions) ingest data from the IELTS Buddy API as described in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or instructions to prevent the agent from interpreting instructions potentially embedded within the retrieved reading materials.
Capability inventory: The skill executes shell commands using the scripts/ielts_buddy_api.py script, providing a mechanism for the agent to perform API calls based on the ingested data.
Sanitization: There is no mention of sanitization or filtering of the external data before it is processed by the agent.
[COMMAND_EXECUTION]: The skill operates by having the agent execute a local Python script (scripts/ielts_buddy_api.py) via the shell to perform all actions, including account binding and data retrieval.
Evidence: Multiple examples in SKILL.md and references/setup.md demonstrate calling the script with various arguments, such as python3 scripts/ielts_buddy_api.py call ielts_review_read_snapshot.
[DATA_EXPOSURE]: The skill interacts with an external API at https://work.ieltsbuddy.igopx.cn/api/v1/agent. This domain is associated with the skill's vendor and is used for legitimate data retrieval purposes.
Credential management: The skill utilizes an environment variable IELTS_BUDDY_TOKEN for authentication. The documentation in references/setup.md explicitly advises against hardcoding secrets and suggests using environment variables or secrets, which aligns with security best practices.

ielts-reading-review