The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill retrieves text from external, untrusted sources (Twitter/X) which could contain malicious instructions designed to influence the agent's behavior when it later reads the generated files.
Ingestion points: Remote content is fetched from api.fxtwitter.com and r.jina.ai across all scripts (fetch_article.py, fetch_tweet.py, and fetch_tweets.sh).
Boundary markers: While generated Markdown files utilize YAML frontmatter for metadata, the retrieved tweet and article content is appended to the body without delimiters or instructions for the agent to ignore embedded commands.
Capability inventory: The skill possesses the capability to execute shell commands (curl, uv) and perform filesystem operations (creating directories and writing Markdown/image files).
Sanitization: Output filenames are sanitized for legal characters, but the retrieved post content is written directly to the filesystem without any sanitization or escaping of potential prompt injection sequences.
[COMMAND_EXECUTION]: The skill uses shell commands to perform its tasks, such as fetching data and executing helper tools.
Evidence: scripts/fetch_article.py and scripts/fetch_tweet.py use subprocess.run to execute curl and uv. The fetch_tweets.sh script executes curl directly.
Context: The implementation in the Python scripts follows best practices by passing command arguments as a list rather than a single string, which effectively mitigates shell injection vulnerabilities.
[EXTERNAL_DOWNLOADS]: The skill relies on several external services and dynamic package installations to function.
Evidence: It downloads tweet data and media from api.fxtwitter.com, r.jina.ai, and pbs.twimg.com. It also uses uv to dynamically fetch the twitter-cli and pyyaml packages at runtime.
Context: The skill utilizes well-known services for Twitter scraping and standard package registries, which is consistent with its primary purpose.

twitter-reader