The Agent Skills Directory

[COMMAND_EXECUTION]: The skill generates and executes a shell script named run-acceptance-tests.sh to perform automated acceptance testing.\n- [COMMAND_EXECUTION]: The agent is instructed to execute project-specific unit test commands and specialized tools like /atdd:mutate and /atdd:kill-mutants.\n- [COMMAND_EXECUTION]: The skill reads from the ~/.claude/teams/ directory to manage and identify existing agent team environments.\n- [PROMPT_INJECTION]: The skill handles untrusted data from feature descriptions which are processed into executable specifications, creating an indirect prompt injection surface.\n
Ingestion points: The skill reads specifications from local specs/*.txt files which are generated from user descriptions.\n
Boundary markers: There are no explicit markers or instructions used to delimit specification content from agent instructions during processing.\n
Capability inventory: The skill possesses the ability to execute shell scripts and modify project code through mutation testing tools.\n
Sanitization: The workflow incorporates a 'Spec Review' phase where a dedicated reviewer agent audits the specifications for implementation leakage.

atdd-team