The Agent Skills Directory

[PROMPT_INJECTION]: Indirect prompt injection vulnerability surface detected.
Ingestion points: Untrusted content is retrieved from the Twitter API and processed in scripts/fetch_tweets.py and scripts/fetch_single_tweet.py.
Boundary markers: The skill uses Markdown blockquotes (>) to delimit tweet content in its output, but does not provide explicit instructions to the agent to ignore any commands contained within the fetched text.
Capability inventory: The skill is granted access to powerful tools including Bash, Read, Write, and Edit.
Sanitization: There is no evidence of sanitization or filtering of the tweet text to remove potentially malicious instructions before it is presented to the agent.
[DATA_EXFILTRATION]: Access to sensitive credential files on the system.
The scripts are configured to read a Twitter auth_token from a hardcoded path: ~/Documents/trend-crawler-master/trend-crawler/config.yaml.
While this is documented for the skill's primary functionality, accessing sensitive configuration files from a separate project directory represents a data exposure risk.
[EXTERNAL_DOWNLOADS]: Reliance on third-party libraries and local environment configuration.
The skill depends on the tweety (or tweety-ns) and pyyaml Python packages.
scripts/fetch_tweets.py contains logic to resolve dependencies from a specific virtual environment path on the user's filesystem, which may lead to runtime errors or unexpected behavior if the environment is not correctly configured.

twitter-crawler