Generate sandbox policy YAML from Sigil scan results.
npx skills add https://github.com/NOMARJ/sigil --skill generate-policy-nomarj
Please help me install this Agent Skill. Skill: generate-policy Source: https://github.com/NOMARJ/sigil/tree/main/plugins/claude-code/skills/generate-policy Command: npx skills add https://github.com/NOMARJ/sigil --skill generate-policy-nomarj
Automates turning Sigil scan findings into a concrete sandbox policy YAML that constrains filesystem, network, process, and credential access so untrusted agent code can be executed safely.
Generate a sigil sandbox policy for the project at ./project-path using the latest scan results and save it to sigil-policy.yaml for review.
High-intent search queries and answers about installing and using this skill.
To generate a sandbox policy YAML from security scan results, map scan findings to filesystem, network, process, and credential restrictions using severity-based mappings and output a ready-to-review sigil-policy.yaml file.
Restrict filesystem and network access for untrusted agent code by synthesizing sandbox policies from scan verdicts, applying deny-all rules for critical findings and permissive rules for clean results to enforce safe execution.
Severity-based enforcement in agent sandboxing applies different restriction levels based on scan verdicts, mapping CLEAN findings to permissive access and CRITICAL findings to deny-all policies for filesystem, network, process, and credential controls.
Yes, you can use sandbox policy generation for CI workflows and third-party skills by producing a YAML policy containing filesystem, network, process, and credentials sections with severity-based mappings before allowing runtime execution.
Make sandbox policies explainable for auditing by including human-readable explanations tying each filesystem, network, process, and credential restriction to specific scan findings to support review.
Automated policy synthesis from scan findings relies on severity-based mappings and requires human review of the generated sigil-policy.yaml to verify that filesystem, network, process, and credential restrictions match the intended enforcement scope.
NOMARJScan repositories for malicious patterns and credential exposures using Sigil CLI.
latchagentSandbox macOS agent shells with filesystem and network restrictions.
alexica00Inspect Claude Code sandbox status, configuration, and security events.
UseAI-proGenerate Docker sandbox configurations with filesystem, network, and process restrictions.
