The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted content from bug reports and audits to drive the patch preparation and implementation process.
Ingestion points: The skill reads data from bug report files in production/qa/bugs/*.md and security audit files in production/security/security-audit-*.md to load release context and scope fixes.
Boundary markers: The skill does not define clear boundary markers or instructions to ignore embedded commands when interpolating bug report content into the tasks assigned to sub-agents.
Capability inventory: The skill utilizes Bash, Edit, and Write tools, and spawns sub-agents (lead-programmer, qa-tester) which have the capability to modify source code and execute system commands.
Sanitization: No validation or sanitization is performed on the ingested bug reports before they are used to define the implementation scope and technical context for sub-agents.
[COMMAND_EXECUTION]: The skill uses the Bash tool to perform verification tasks, including running /smoke-check and executing automated unit and integration tests found in the tests/ directory.

day-one-patch