The Agent Skills Directory

[COMMAND_EXECUTION]: The skill facilitates interaction with the TAPD platform by executing a local Python script (scripts/tapd.py). The instructions guide the agent to construct and run shell commands to perform operations like querying, creating, and updating project entities.
[SAFE]: The skill implements a secure approach to authentication by requiring users to set environment variables such as TAPD_ACCESS_TOKEN or TAPD_API_USER/TAPD_API_PASSWORD. This avoids hardcoding sensitive credentials within the skill itself. Additionally, all network operations target well-known official TAPD domains (api.tapd.cn).
[PROMPT_INJECTION]: The skill ingests data from external, user-controllable sources on the TAPD platform (e.g., requirement descriptions, comments, and Wiki pages). This data is processed by the agent, creating a surface for potential indirect prompt injection attacks.
Ingestion points: Data enters the agent's context through commands like get_stories_or_tasks, get_comments, and get_wiki as defined in SKILL.md.
Boundary markers: The skill documentation does not define specific delimiters or instructions to the agent to treat fetched content as untrusted data.
Capability inventory: The agent possesses the capability to execute shell commands (tapd.py) and perform write operations to the TAPD platform.
Sanitization: There is no description of content sanitization or validation for the data retrieved from external TAPD sources.

tapd