The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes output from external tool commands.
Ingestion points: Data is read from the gplay testers get command and stored in the $CURRENT variable in SKILL.md.
Boundary markers: None identified; the skill does not use delimiters or warnings to separate tool output from instructions.
Capability inventory: The skill executes gplay commands and shell utilities including jq, grep, and paste.
Sanitization: The skill uses jq and grep for data extraction and filtering, but does not perform sanitization to prevent the interpolation of malicious content from the tool output into subsequent commands.
[COMMAND_EXECUTION]: The skill generates shell commands that incorporate data retrieved from external sources.
Evidence: Use of jq to parse JSON results and grep/paste to construct email lists for the gplay testers update command.

gplay-testers-orchestration