The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill mandates a workflow involving external data ingestion which presents a vulnerability surface for indirect prompt injection.
Ingestion points: The mandatory workflow in SKILL.md requires spawning agents to perform web research (fuse-ai-pilot:research-expert via Context7/Exa) and query external documentation (mcp__context7__query-docs).
Boundary markers: Absent. The instructions do not provide delimiters or warnings to ignore embedded instructions within the data retrieved during the research phase.
Capability inventory: The skill utilizes shell execution for project initialization, package installation, and tool configuration (npm, npx, biome, eslint) as documented in references/biome-setup.md and references/eslint-typed.md.
Sanitization: Absent. There is no evidence of content filtering or validation for the data retrieved from external research before it influences the agent's actions.
[EXTERNAL_DOWNLOADS]: The skill references and fetches resources from official, well-known technology providers.
Fetches the official Biome JSON schema from biomejs.dev in references/templates/config-examples.md.
Directs agents to consult documentation from official sources including biomejs.dev and typescript-eslint.io.

ts-lint-format