The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent ingests and acts upon external data sources, creating a potential surface for indirect injection attacks.
Ingestion points: The agent is instructed to read domain language documentation, Architecture Decision Records (ADRs), and the output of executed test commands in the SKILL.md file.
Boundary markers: No explicit delimiters or instructions to ignore embedded prompts within the domain language or test outputs are defined.
Capability inventory: The skill possesses the capability to write to the file system (creating code and tests) and execute shell commands (running the test suite) via the wiki forge tool suite.
Sanitization: No sanitization or validation logic is present to filter potentially malicious instructions embedded in the external data before processing.
[COMMAND_EXECUTION]: The skill workflow relies on the execution of shell commands to verify code behavior.
Evidence: The SKILL.md file explicitly instructs the agent to run commands such as wiki forge tdd red, wiki forge tdd green, and wiki forge tdd cycle to progress through the development cycle. These commands incorporate user-provided or agent-generated strings as arguments for test execution.

tdd