The Agent Skills Directory

[DYNAMIC_EXECUTION]: The script scripts/fingerprint_storage.py performs dynamic path injection to import the StorageAPI module. It uses sys.path.insert with a path constructed at runtime based on the file's relative location to load code from a sibling skill directory (results-storage).
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data by fetching HTTP headers, cookies, and page content from target URLs, which could be used to influence the agent's behavior.
Ingestion points: scripts/extract_headers.py, scripts/tech_matcher.py, and scripts/waf_detector.py fetch and analyze data from external web responses.
Boundary markers: The skill lacks explicit boundary markers or instructions to the agent to ignore potentially malicious commands embedded in the fetched content.
Capability inventory: The skill possesses capabilities for network requests, shell command execution (via external tools), and local file/database storage.
Sanitization: While the scripts use regular expressions for technology matching, they do not include robust sanitization to prevent an LLM from interpreting instructions contained within the analyzed web content.
[EXTERNAL_DOWNLOADS]: The SKILL.md file and various guides within the skill provide instructions for the user to download and install several third-party security tools from external sources, including github.com/projectdiscovery/httpx, github.com/projectdiscovery/nuclei, github.com/praetorian-inc/fingerprintx, and wafw00f via Python's package registry.
[COMMAND_EXECUTION]: The skill relies extensively on executing external command-line tools such as curl, whatweb, nmap, wafw00f, nuclei, and httpx to perform its primary fingerprinting functions.

recon-fingerprint