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). High: the required runtime workflow fetches outsider-authored free text from public web sources (X HTML/JS and tweet/article content via fetchHomeHtml/fetchText/GraphQL response.text()), which is then parsed and incorporated into generated markdown and thus into the agent’s LLM context if the agent reads/returns that markdown.

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

• Secret detected (high risk: 1.00). I scanned the repository for high-entropy literal values that could provide access.
• Flagged: scripts/constants.ts DEFAULT_BEARER_TOKEN (lines 3-4) — this is a long, high-entropy "Bearer ..." token baked into code and used as a fallback X/Twitter bearer token. It appears to be a real credential-like value and can be used for API access, so it qualifies as a secret.
• Ignored items and why:
• Various query IDs and feature switch strings (e.g., FALLBACK_QUERY_ID, FALLBACK_TWEET_QUERY_ID) — these are API/query identifiers, not credentials that grant access.
• Cookie name lists and environment variable names (X_AUTH_TOKEN, X_CT0, etc.) — just names/placeholders; no secret values present.
• DEFAULT_USER_AGENT — harmless user-agent string, not a secret.
• bun.lock sha512 entry — package integrity hash, not a credential.
• Other long-looking strings (e.g., GraphQL IDs) are implementation identifiers, not authentication tokens.

Therefore the only actionable secret present is the DEFAULT_BEARER_TOKEN literal.