attack
Route an attack vector to the appropriate specialist agent — usage: /project:attack <target> <vector>
Start a new engagement for a target — verifies scope, creates evidence directories, and launches recon agent
> /plugin marketplace add mukul975/Threatswarm > /plugin install threatswarm@threatswarm
How it fires
How this command gets triggered: by you, by Claude, or both.
/engageContext preview
What this command does when you run it.
Start a new engagement for a target — verifies scope, creates evidence directories, and launches recon agent
description: Start a new engagement for a target — verifies scope, creates evidence directories, and launches recon agent allowed-tools: Bash, Read, Write, Glob
Start a new penetration testing engagement for target: $ARGUMENTS
Follow these steps in order:
1. **Scope Verification**: Read `scope.txt` and verify that "$ARGUMENTS" is listed as an authorized target (exact IP, CIDR that includes the IP, or domain match). If NOT found in scope.txt, STOP immediately and output:
❌ SCOPE VIOLATION: "$ARGUMENTS" is not in scope.txt Add the target to scope.txt before proceeding.
2. **Create Evidence Directories**: Create the full directory structure:
mkdir -p evidence/$(date +%Y%m%d)/$ARGUMENTS/{nmap,nuclei,web,creds,screenshots,post_ex,ad,cloud,wireless,mobile,api,re,osint,network,lateral,logs}3. **Print Engagement Brief**:
═══════════════════════════════════════════════ ENGAGEMENT START Target: $ARGUMENTS Date: $(date -u +%Y-%m-%dT%H:%M:%SZ) Operator: $(whoami) Scope: VERIFIED ✓ Evidence: evidence/$(date +%Y%m%d)/$ARGUMENTS/ ═══════════════════════════════════════════════
4. **Delegate to Recon Agent**: Invoke the `recon` sub-agent with the target. The recon agent will:
5. **Parse Recon Results**: After recon completes, read `recon_summary.md` and extract:
6. **Print Attack Vector Recommendations**: Based on the recon findings, output a prioritized list:
══ RECOMMENDED ATTACK VECTORS ══════════════════ Priority 1: [e.g., "CVE-XXXX on Apache 2.4.49 (port 80)"] Priority 2: [e.g., "Default credentials on admin panel (/admin)"] Priority 3: [e.g., "SQL injection on login form"] Priority 4: [e.g., "Outdated OpenSSH — check for user enumeration"] ════════════════════════════════════════════════ Ready. Run: /project:attack $ARGUMENTS <vector>
27 scope-enforced AI agents that run the full pentest kill-chain (recon → exploit → post-ex → DFIR → report) as a one-command Claude Code plugin. Backed by 754 MITRE-mapped skills.
Repo: mukul975/Threatswarm
Route an attack vector to the appropriate specialist agent — usage: /project:attack <target> <vector>
Incident response workflow — triage, evidence collection, timeline, and IOC extraction