The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes product information (titles, descriptions, categories) from TikTok via the Kalodata API. This data is ingested into the agent's context without explicit boundary markers or sanitization, creating a potential surface for indirect prompt injection if malicious instructions are embedded in product metadata.
Ingestion points: API responses processed and displayed by scripts/kalodata_product_search.py and scripts/kalodata_product_detail.py.
Boundary markers: Absent in both scripts and documentation.
Capability inventory: The skill possesses the capability to write files to the local project directory and perform network operations to vendor-specific domains.
Sanitization: No sanitization or escaping of external product content was detected in the provided scripts.
[PERSISTENCE]: The skill's onboarding documentation (references/onboarding.md) instructs users to modify their shell profiles (~/.zshrc, ~/.bashrc) to persist the LINKFOX_AGENT_API_KEY environment variable. While this is a common practice for CLI tool configuration, it represents a persistent modification to the user's environment configuration.
[DATA_EXFILTRATION]: The skill's search and detail scripts transmit session-related metadata, including SESSION_ID, MESSAGE_ID, MODE_ID, and APP_NAME, as HTTP headers to the vendor's API gateway (tool-gateway.linkfox.com). Additionally, the scripts/onboarding.py utility communicates with several vendor-owned domains (api.linkfox.com, agent-api.linkfox.com, skill-api.linkfox.com) to manage user authentication, account registration, and billing. These operations are consistent with the skill's stated purpose and vendor infrastructure.

linkfox-kalodata-tiktok-product