The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the ae-cli command-line tool to manage project configurations, fetch tracking plans, and add file attachments. These operations are performed within the scope of the skill's purpose to facilitate data tracking integration.
[EXTERNAL_DOWNLOADS]: The skill fetches tracking plan data from the user-configured ThinkingData host via ae-cli tracking plan fetch. It also provides links to official vendor documentation and software downloads (LogBus2) hosted on the vendor's official domain (thinkingdata.cn).
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests untrusted data from tracking plans (.xlsx or remote plans) and interpolates this data into code generated via the Edit tool.
Ingestion points: .ae-cli/draft.xlsx and remote plan JSON fetched via ae-cli.
Boundary markers: The instructions specify that event and property names should be mapped strictly from the plan.
Capability inventory: Uses the Edit tool to modify project source code and ae-cli for environment management.
Sanitization: The skill relies on predefined templates for code generation, though it lacks explicit validation rules for user-supplied property or event names within those templates.
[SAFE]: The skill incorporates safety checks, specifically requiring that the user's git workspace is clean before performing any automated code insertions to prevent accidental data loss or unreviewed changes.

ae-generate-tracking-code