The Agent Skills Directory

[PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill processes untrusted content retrieved from the TAPD platform (e.g., Wiki pages, Story descriptions, Bug titles) which could contain malicious instructions.
Ingestion points: scripts/search_wiki.py downloads content from the TAPD API and stores it in local files.
Boundary markers: Absent. The skill does not use specific delimiters or instructions to treat the fetched data as untrusted.
Capability inventory: The skill is granted access to the Bash, Read, and Glob tools.
Sanitization: Absent. Data from the API is saved and read without validation or escaping.
[COMMAND_EXECUTION]: The skill relies on the execution of a local Python script (scripts/search_wiki.py) and various shell commands (curl, ls, cat) to interact with the TAPD API and manage local cache files.
[DATA_EXFILTRATION]: The skill manages a sensitive authentication token (TAPD_TOKEN) through environment variables. This token is included in the headers of curl requests sent to the configured ${TAPD_API_ENDPOINT}. While this is the intended functionality, a user-controlled or compromised API endpoint could lead to credential exposure.

tapd-openapi