The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes multiple shell commands to perform network discovery and service identification.
Evidence: Use of nmap, curl, rpcclient, smbclient, and nc to probe targets for specific SCCM-related ports and services (SMB 445, RPC 135, MSSQL 1433, etc.).
[PRIVILEGE_ESCALATION]: The skill provides instructions to run commands with administrative privileges during the setup phase.
Evidence: The 'Tooling Setup' section includes sudo apt-get install -y nmap curl netcat-openbsd smbclient rpcbind to install necessary utilities.
[EXTERNAL_DOWNLOADS]: The skill facilitates the download and installation of external software packages.
Evidence: Uses apt-get update and apt-get install to fetch system utilities from remote package registries.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from network targets, creating a potential surface for indirect injection via manipulated service responses.
Ingestion points: The agent reads HTTP headers and WMI paths from remote servers via curl and rpcclient.
Boundary markers: None identified in the provided instructions for delimiting remote data.
Capability inventory: The skill has the ability to execute network probes and file system operations via standard CLI tools.
Sanitization: No explicit evidence of output sanitization or validation of data retrieved from the network targets.

sccm-recon