The Agent Skills Directory

[DATA_EXFILTRATION]: The skill is instructed to access and read sensitive files such as .env and infrastructure configuration files (e.g., wrangler.jsonc, vercel.json) to extract metadata and identifiers like Cloudflare account IDs, Storyblok space IDs, and environment variable names used for authentication. While the intended use is to populate documentation, accessing .env files is a sensitive operation that exposes potential secrets to the agent's context.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the repository's file system to generate the backstage.yaml manifest.
Ingestion points: The agent reads from various repository files, including source code, package.json, .env files, and deployment configurations using Read, Glob, and Grep tools.
Boundary markers: No specific delimiters or instructions to ignore embedded commands within the ingested files are provided.
Capability inventory: The skill possesses Write and AskUserQuestion capabilities, allowing it to modify the repository based on the parsed data.
Sanitization: There are no explicit sanitization or validation steps mentioned to filter malicious instructions that might be embedded in the repository files being analyzed.

backstage