metasploit-cursor-harness: Agent for Claude Code

.cursor/agents/harness-orchestrator.md

harness-orchestrator is an agent for Claude Code, Cursor from Suzu-Testing/metasploit-cursor-harness. It costs 53 tokens per session (1,597 once invoked), scanned A, original, MIT.

An orchestrator for Metasploit penetration-testing engagements, which are authorised security tests of systems. It coordinates testing phases, checkpoints, and specialist agents.

In plain words
What is it for?
Use it to start reconnaissance, testing, or exploitation workflows, track engagement state, and coordinate the agents handling each phase.
Why use it?
It helps keep broad security assessments in the required order and prevents work from moving past missing scope, approvals, or checks.

Agent for Claude CodeCursor

Written for Cursor and Claude Code: installed under .cursor/, but also a Claude Code subagent (agents/*.md). Also seen: model in frontmatter; mentions subagents.

This is Suzu-Testing/metasploit-cursor-harness's own configuration. It tells Claude Code and Cursor how to work on metasploit-cursor-harness itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything metasploit-cursor-harness configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is - Run: python .cursor/skills/pentest-workflow/scripts/complete-subgate.py {subgate_id} --engagement {engagement_id} --evidence {path}.

Reuse

Borrowing it

Nothing to install: this file belongs to Suzu-Testing/metasploit-cursor-harness. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/Suzu-Testing/metasploit-cursor-harness/main/.cursor/agents/harness-orchestrator.md
Clone the repo
git clone --depth 1 https://github.com/Suzu-Testing/metasploit-cursor-harness

Made for: Claude Code, Cursor.

Wrote 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.

agentmods badge for harness-orchestrator

README.md
[![agentmods](https://agentmods.dev/badge/agents/suzu-testing/metasploit-cursor-harness/harness-orchestrator.svg)](https://agentmods.dev/agents/suzu-testing/metasploit-cursor-harness/harness-orchestrator)
Your own site
<a href="https://agentmods.dev/agents/suzu-testing/metasploit-cursor-harness/harness-orchestrator"><img src="https://agentmods.dev/badge/agents/suzu-testing/metasploit-cursor-harness/harness-orchestrator.svg" alt="Measured on agentmods" height="20"></a>
Per session 53 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,597 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00053 $0.01597
Opus 5 $0.00026 $0.00798
Sonnet 5 $0.00011 $0.00319
Haiku 4.5 $0.00005 $0.00160

Measured 6d ago against content hash 97eaee44ae99, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

harness-orchestrator 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 6d 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.

.cursor/agents/harness-orchestrator.md · 153 lines

How it starts

The opening of the file, as written. The whole thing — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the master coordinator for Metasploit-based penetration testing engagements through the msf-harness MCP server.

Your Mission

Drive the phase -> gate -> subgate workflow. Read engagement state, complete or delegate subgates, validate gates, and advance phases. You coordinate; specialized agents execute.

Preconditions (MUST verify before any action)

  1. Resolve engagement_id:
    • IF user specifies one -> use it
    • ELSE IF only one directory exists in engagements/ (excluding _template, catalogs) -> use it
    • ELSE -> ask user; do NOT guess
  2. Run python .cursor/skills/pentest-workflow/scripts/gate-check.py {engagement_id}
  3. IF exit code != 0 AND output shows GATE STATUS: BLOCKED -> report blocked items; do NOT proceed past the gate
  4. Read these files (ALL mandatory before first action):
    • engagements/{engagement_id}/phase-state.yaml
    • engagements/{engagement_id}/roe.yaml
    • engagements/catalogs/phases.yaml
    • scope/scope-master.txt

IF any file is missing -> STOP and report which file is absent.

Workflow Loop (strict order)

1. gate-check.py {engagement_id}
   - Record: current_phase, pending_subgates[], gate_status
   - IF gate_status == PASS -> run gate-check.py {engagement_id} --advance; re-run step 1
   - IF gate_status == BLOCKED -> proceed to step 2

2. Select FIRST pending subgate (by catalog order)
   - Look up subgate in delegation table below
   - IF subgate.agent == "self" -> execute using subgate.skill
   - IF subgate.agent != "self" -> delegate ONCE to that agent

3. On delegation return OR self-execution complete:
   - VERIFY: evidence file exists at reported path
   - IF evidence missing -> do NOT complete subgate; report BLOCKED
   - Run: python .cursor/skills/pentest-workflow/scripts/complete-subgate.py {subgate_id} --engagement {engagement_id} --evidence {path}
   - IF script fails -> STOP and report error

4. Re-run gate-check.py {engagement_id}
   - IF more pending subgates -> go to step 2
   - IF gate passes -> run --advance; go to step 1
   - IF gate blocked on non-subgate criteria -> report and STOP

5. STOP when: current_phase == "complete" OR user requests stop

Read the full file on GitHub · 153 lines

Changes

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.

  1. 6d ago First seen · 153 lines · 53 tokens per session scan A 97eaee44ae99

Subscribe to this mod's changes

harness-orchestrator is an agent published in the GitHub repository Suzu-Testing/metasploit-cursor-harness (3 stars, last pushed 13d ago), licensed MIT. It adds 53 tokens to every session and 1,597 once invoked, about $0.0003 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.