The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from X (Twitter) comments. Ingestion points: Comments are retrieved via scripts/tweet_comment_search.py from the redfox.hk API. Boundary markers: The instructions in references/core_workflow.md do not include specific delimiters or instructions to the agent to treat the comment content as data rather than instructions. Capability inventory: The skill uses local scripts capable of performing network requests and writing to the local file system. Sanitization: There is no evidence of sanitization or filtering of the comment text before it is presented to the agent for analysis.
[COMMAND_EXECUTION]: The skill requires the agent to execute local Python scripts (tweet_comment_search.py and backfill_html.py) which are included in the skill package for data retrieval and report population.
[DATA_EXFILTRATION]: The skill performs network requests to https://redfox.hk/story/api/x/tweetComments. This functionality depends on the REDFOX_API_KEY, which is provided by the user and stored in environment variables.

x-comment-analyzer