The Agent Skills Directory

[SAFE]: The skill consists of instructional documentation and code examples focused on the secure use of Go's os/exec package. No malicious patterns or security risks were identified.
[PROMPT_INJECTION]: The content does not contain instructions to override agent behavior or bypass safety guidelines. All instructions are relevant to the primary purpose of Go process supervision.
[DATA_EXFILTRATION]: Analysis found no evidence of sensitive data harvesting or unauthorized network transmission. Guidelines recommend redacting credentials from diagnostic output.
[COMMAND_EXECUTION]: The skill provides best practices for executing subprocesses, such as separating executable names from arguments to prevent shell injection vulnerabilities.
[INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing potentially untrusted data from external subprocesses.
Ingestion points: Processes output via Cmd.Stdout and Cmd.Stderr as described in references/process-control/knowledge.md.
Boundary markers: Encourages explicit status checks and output predicates in references/process-control/rules.md.
Capability inventory: Demonstrates execution of external binaries using exec.CommandContext in references/process-control/examples.md.
Sanitization: Recommends redacting credentials and tokens from captured output in references/process-control/rules.md.

go-cli-process-control