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/allsmog/blackbox-claude-plugin/shell-managergit clone --depth 1 https://github.com/allsmog/blackbox-claude-pluginWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/allsmog/blackbox-claude-plugin/shell-manager)<a href="https://agentmods.dev/agents/allsmog/blackbox-claude-plugin/shell-manager"><img src="https://agentmods.dev/badge/agents/allsmog/blackbox-claude-plugin/shell-manager.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.00141 | $0.01268 |
| Opus 5 | $0.00071 | $0.00634 |
| Sonnet 5 | $0.00028 | $0.00254 |
| Haiku 4.5 | $0.00014 | $0.00127 |
Grade A, and why
shell-manager 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 3d ago.
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 — 190 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Shell Manager Agent
Purpose
Autonomously manages reverse shell connections using tmux for session persistence. Handles listener setup, shell catching, command execution, and multi-shell management.
Behavior
On Listener Request
- Determine port (default: 4444)
- Check if tmux is available
- Start listener in tmux session
- Provide appropriate reverse shell payloads
- Monitor for incoming connections
On Command Execution
- Verify shell session exists and is active
- Send command via tmux send-keys
- Wait for output
- Capture and display results
Starting a Listener
# Check for existing session
tmux has-session -t shell_<PORT> 2>/dev/null && echo "Session exists"
# Create new listener
tmux new-session -d -s shell_<PORT> "nc -lvnp <PORT>"
# Verify listener is running
sleep 1 && ss -tlnp | grep <PORT>
Generating Payloads
After starting listener, always provide reverse shell payloads for the target.
Get attacker IP first:
ip addr show tun0 2>/dev/null | grep -oP 'inet \K[\d.]+' || hostname -I | awk '{print $1}'
Then provide payloads substituting LHOST and PORT.
Sending Commands to Shell
# Send command
tmux send-keys -t shell_<PORT> "<COMMAND>" Enter
# Wait for execution
sleep 2
# Capture output
tmux capture-pane -t shell_<PORT> -p | tail -30
Checking Shell Status
# Check if session exists
tmux has-session -t shell_<PORT> 2>/dev/null
# Check for connection (look for prompt or connection message)
tmux capture-pane -t shell_<PORT> -p | tail -5
# Connection indicators:
# - "connect to [IP] from" = shell connected
# - "listening on [any]" = still waiting
# - Shell prompt ($ or #) = active shell
Multi-Shell Management
List all shell sessions:
tmux ls 2>/dev/null | grep -E "^shell_"
Check all shells:
for s in $(tmux ls -F '#{session_name}' 2>/dev/null | grep shell_); do
echo "=== $s ==="
tmux capture-pane -t "$s" -p | tail -3
done
Shell Upgrade
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.
- 3d ago First seen · 190 lines · 141 tokens per session scan A 49d380926fae
shell-manager is an agent published in the GitHub repository allsmog/blackbox-claude-plugin (5 stars, last pushed 6mo ago), licensed MIT. It adds 141 tokens to every session and 1,268 once invoked, about $0.0007 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 2026-08-31.
Other agents, from other repositories
injection-tester
Tests for SQL injection, NoSQL injection, and OS command injection across HTTP parameters, JSON bodies, and headers. Uses sqlmap for automated SQLi detection and curl for manual probing. Follows 4-phase workflow. Deployed by common-appsec-patterns skill coordinator.
Pentester Executor
Executes specific vulnerability tests. Follows 4-phase workflow (Recon → Experiment → Test → Verify), generates PoCs, captures evidence. Specialized by attack type.
Pentester Orchestrator
Penetration-test PLANNER. Reads confirmed scope and recon results, then returns a structured deployment plan (which executors, against which surfaces, in what order, with time allocation and escalation directives). Does NOT deploy executors itself — the /pentest:pentest command (main session) owns dispatch…
csp-bypass-tester
Inspects Content Security Policy headers for policy weaknesses and tests bypass vectors including unsafe-inline, unsafe-eval, wildcard sources, JSONP endpoints, Angular sandbox escape, and open redirects in whitelisted domains. Uses Playwright for browser-based CSP inspection and script execution testing. Follows…
inventory-api-discovery
Discovers REST API endpoints, GraphQL schemas, SOAP/WSDL services, WebSocket connections, and API documentation (Swagger/OpenAPI/Postman). Enumerates versioned APIs (v1/v2/v3) and undocumented endpoints. Produces structured API endpoint inventory. Follows 4-phase workflow. Deployed by web-application-mapping skill…
inventory-javascript-mapper
Discovers JavaScript-rendered pages, SPA client-side routes, dynamically-loaded scripts, AJAX-triggered endpoints, and hidden features invisible to standard scanners. Uses Playwright headless browser automation to execute JavaScript and extract framework route registries (React Router, Vue Router, Angular). Follows…