The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use curl and shell piping to perform reconnaissance and automated probing of external website infrastructures.
[DATA_EXFILTRATION]: The instructions explicitly guide the agent to locate and extract sensitive third-party data, including Google Maps API keys (detected via regex AIza[a-zA-Z0-9_-]{35,}) and customer PII (names, addresses, phone numbers) found in exposed /wp-content/debug.log files and CRM portals.
[PROMPT_INJECTION]: Vulnerability to Indirect Prompt Injection (Category 8). The skill fetches and processes untrusted data from external websites using curl without sanitization, allowing content on the target sites to potentially influence the agent's logic or instructions.
Ingestion points: Website responses and HTML content fetched via curl in the Step-by-Step section of SKILL.md.
Boundary markers: None present; external content is piped directly into analysis tools like grep and head.
Capability inventory: Execution of shell commands via curl and writing temporary files to /tmp/estimate.html.
Sanitization: No escaping or validation is performed on the data retrieved from external URLs before processing.

recon-tree-services