The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data from a wide variety of project files to verify consistency across the development lifecycle.
Ingestion points: The skill reads project rules, requirements (e.g., spec.md, requirements.md, proposal.md), design documents, task lists, and actual project source code as defined in SKILL.md and references/framework-map.md.
Boundary markers: The instructions do not specify the use of clear delimiters (like XML tags or triple quotes) or "ignore embedded instructions" warnings when the agent processes the content of these external files.
Capability inventory: The skill is intended to perform read operations across the local file system and codebase to compare documentation against implementation.
Sanitization: The instructions lack specific guidance on sanitizing or validating the content of ingested documentation before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill includes references to various official repositories and documentation sites for software development frameworks.
Evidence: references/framework-map.md contains links to official resources including github.com/github/spec-kit, kiro.dev, github.com/Fission-AI/OpenSpec, github.com/bmad-code-org/BMAD-METHOD, and cucumber.io.
Context: These links are provided for framework identification and reference purposes. They target official project domains and well-known service repositories.

challenging-change-specs