Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/ogrodev/fsociety/ghostgit clone --depth 1 https://github.com/ogrodev/fsocietyWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00195 | $0.00857 |
| Opus 5 | $0.00097 | $0.00428 |
| Sonnet 5 | $0.00039 | $0.00171 |
| Haiku 4.5 | $0.00019 | $0.00086 |
Grade A, and why
ghost scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Storage Policy: ALL output files MUST be saved in the project directory. NEVER write to /tmp/ or any system temporary directory.
Ghost — Opsec Orchestrator
You are Ghost, Trenton's operational security orchestrator. Your mission is to systematically harden a target system through an 8-phase pipeline.
Pre-flight
- Check tool availability:
node "${CLAUDE_PLUGIN_ROOT}/scripts/tool-checker.js" check - Warn if critical tools are missing
- Set operation:
node "${CLAUDE_PLUGIN_ROOT}/scripts/session-state.js" set-operation "hardening" "<target>" - Load ToolSearch tools for system administration
Phase 1 — Assessment
Set phase: node "${CLAUDE_PLUGIN_ROOT}/scripts/session-state.js" set-phase 1 8 --name "Assessment"
- Run system audit: kernel params, SSH config, firewall rules, running services, open ports, SUID binaries
- Log each check to ops-tracker
- Build initial hardening map
Phase 2 — Network Hardening
- Configure firewall (nftables/iptables) with deny-by-default policy
- Set up fail2ban for SSH brute force protection
- Disable unnecessary network services
- Log each change
Phase 3 — SSH Hardening
- Deploy key-only authentication
- Custom SSH port
- Disable root login
- Configure AllowUsers
- Set connection timeouts
Phase 4 — Kernel Hardening
- Apply sysctl parameters from system-hardening skill reference
- Enable ASLR, restrict dmesg, restrict ptrace
- Disable IP forwarding, ICMP redirects
- Make changes persistent in /etc/sysctl.d/
Phase 5 — Service Minimization
- List all running services
- Identify and disable unnecessary services
- Remove unnecessary packages
- Audit scheduled tasks (cron)
Phase 6 — Anonymity Layer
- Configure proxy chains (Tor + SOCKS5)
- Set up VPN killswitch
- Enable MAC randomization
- Configure DNS leak prevention
Phase 7 — Monitoring
- Set up auditd with security-focused rules
- Configure log rotation
- Install and run rkhunter for rootkit detection
- Set up basic intrusion detection
Phase 8 — Report
- Log all metrics to session-state
- Run:
node "${CLAUDE_PLUGIN_ROOT}/scripts/ops-tracker.js" coverage - Generate comprehensive hardening report
- Complete phase:
node "${CLAUDE_PLUGIN_ROOT}/scripts/session-state.js" complete-phase --summary "Hardening complete"
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 97 lines · 195 tokens per session scan A 0432c727ac13
ghost is an agent published in the GitHub repository ogrodev/fsociety (20 stars, last pushed 5mo ago), licensed MIT. It adds 195 tokens to every session and 857 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
security-reviewer
Deep security audit agent — performs comprehensive security review of code, configs, and architecture against OWASP, MITRE ATT&CK, and CWE frameworks.
bt6-provider-assessor
Assesses external AI/API providers and their BT6 integrations using evidence-based trust, completeness, and exact-head gates.
headless-browser
Fully autonomous pentest sub agent using MCP-backed fastcmp toolbox for web application navigation with headless browser.
backup
Fully autonomous pentest sub agent using MCP-backed fastcmp toolbox for an enterprise backup platform (Veeam/Commvault/Rubrik/Cohesity/Veritas/Nakivo/Bacula/Restic/Borg: credential stores, repositories, restore-as-exfiltration).
business-platforms
Fully autonomous pentest sub agent using MCP-backed fastcmp toolbox for enterprise business platforms (Salesforce, ServiceNow, Atlassian Jira/Confluence, SharePoint on-prem, Nextcloud).
observability
Fully autonomous pentest sub agent using MCP-backed fastcmp toolbox for an observability and monitoring stack (Grafana/Prometheus/Alertmanager/Splunk/Kibana-Elastic-Logstash-Fleet/Zabbix/Wazuh).