Pular para o conteúdo
TakoAPI
Voltar às skills

fletcher-cyber-security-engineer

por fletcherfrimpong

Coding Agents & IDEs
629Downloads informados pelo clawskills.sh
0Visualizações
0 curtidaEntre para curtir esta skill
—Ainda sem avaliações
Sua avaliação

Resumo

Security engineering workflow for OpenClaw privilege governance and hardening.

Descrição

O que esta skill faz

Enforces least-privilege execution, approval-gated privilege escalation, and session timeout controls within OpenClaw workflows. Monitors listening ports and outbound connections against approved baselines. Generates compliance reports mapped to ISO 27001 and NIST controls with concrete mitigations.

Combines runtime privilege governance, network monitoring, and compliance benchmarking in one workflow instead of managing separate tools for each concern.

Quando usá-la

  • Auditing which ports are exposed before deploying a service
  • Generating a privilege escalation audit trail for a security review
  • Flagging outbound connections to unapproved destinations during a task
  • Running a compliance gap report against ISO 27001 controls
  • Enforcing command allow/deny policy in a sensitive automation workflow

Exemplo de fluxo de trabalho

ENTRADA

User asks: Run a preflight security check on my current environment and report any violations

AGENTE
  1. 1

    Runs preflight_check.py to assess current privilege state and session context

  2. 2

    Invokes port_monitor.py to compare listening ports against the approved baseline

  3. 3

    Runs egress_monitor.py to identify outbound connections outside the allowlist

  4. 4

    Maps findings to ISO 27001 and NIST controls via compliance-controls-map.json

  5. 5

    Calls notify_on_violation.py and outputs a report with check_id, status, risk, and mitigations

SAÍDA

Compliance report listing each violation with affected check_id, risk level, evidence, and specific remediation steps

Instalação

clawhub install fletcherfrimpong/fletcher-cyber-security-engineer

Execute o comando da CLI ou cole o prompt no Claude Code, Cursor, Codex ou no agente de código que você usa.

Links