The Agent Skills Directory

[SAFE]: The skill consists of documentation and configuration files intended to guide AI behavior during coding tasks. There are no command executions, remote downloads, or credential exposures.
[INDIRECT_PROMPT_INJECTION]: The skill directs the agent to ingest project-specific documentation, including GLOSSARY.md and Architecture Decision Records (ADRs). This is a standard coding assistant feature but represents a surface for indirect prompt injection if those files are controlled by an adversary. 1. Ingestion points: GLOSSARY.md and ADR files in the user's codebase (referenced in SKILL.md). 2. Boundary markers: No explicit markers are defined for the ingested files. 3. Capability inventory: The agent is expected to possess file-system access and shell execution capabilities to fulfill the TDD lifecycle. 4. Sanitization: No sanitization or validation of the external content is described.

tdd