baoyu-danger-x-to-markdown

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The skill fetches outsider-authored free text from X at runtime (e.g., tweet/article content and referenced tweet text) via fetchTweetResult/fetchXArticle GraphQL calls, then passes that extracted prose into markdown formatting and ultimately into the agent’s output context; there is no LLM prompt-injection sanitization step shown.

HIGH W008: Secret detected in skill content (API keys, tokens, passwords).

• Secret detected (high risk: 1.00). I flagged one hardcoded secret: the DEFAULT_BEARER_TOKEN string in scripts/constants.ts. It is a long, random-looking "Bearer ..." token (high entropy) and is an actual authorization bearer token embedded in source — this matches the definition of a secret (an API credential that can grant access).

I did NOT flag environment variable names (e.g., X_AUTH_TOKEN, X_CT0, etc.) or obvious non-secret strings such as DEFAULT_USER_AGENT, URLs, query IDs, or package integrity hashes — these are either names/placeholders or non-sensitive configuration values. No PEM/private-key blocks or other high-entropy secrets were found besides the bearer token.