osint-orchestrator

osint-orchestrator is an agent for coding agents from daemon-blockint-tech/Grond. It costs 243 tokens per session (771 once invoked), scanned A, original, MIT.

A coding agent for designing and building the main LangGraph orchestration system for Grond, an OSINT platform. OSINT means gathering information from publicly available sources.

In plain words
What is it for?
Planning agent routing, shared state, parallel passive-recon work, human approval before active scans, and confidence-scored result fusion.
Why use it?
It coordinates specialized agents, controls active-scan authorization, preserves session state, and combines findings without letting one error stop the whole workflow.

Agent

Install

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.

agentmods
npx agentmods add agents/daemon-blockint-tech/grond/osint-orchestrator
Clone the repo
git clone --depth 1 https://github.com/daemon-blockint-tech/Grond

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 osint-orchestrator

README.md
[![agentmods](https://agentmods.dev/badge/agents/daemon-blockint-tech/grond/osint-orchestrator.svg)](https://agentmods.dev/agents/daemon-blockint-tech/grond/osint-orchestrator)
Your own site
<a href="https://agentmods.dev/agents/daemon-blockint-tech/grond/osint-orchestrator"><img src="https://agentmods.dev/badge/agents/daemon-blockint-tech/grond/osint-orchestrator.svg" alt="Measured on agentmods" height="20"></a>
Per session 243 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 771 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00243 $0.00771
Opus 5 $0.00121 $0.00385
Sonnet 5 $0.00049 $0.00154
Haiku 4.5 $0.00024 $0.00077

Measured 4d ago against content hash 7e07f345194f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

osint-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 4d 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.

agents/osint-orchestrator.md · 67 lines

What it actually says

You are the OSINT Orchestrator Agent for the Grond platform — responsible for designing, building, and debugging the LangGraph-based multi-agent coordination system.

Your Core Responsibilities:

  1. Design and implement the OsintState typed state schema shared across all agents
  2. Build the LangGraph StateGraph with correct node wiring, edge conditions, and parallel Send() dispatch
  3. Implement the supervisor planner node that decomposes a high-level target + goal into sub-agent work
  4. Enforce the Human-in-the-Loop (HITL) gate before any active scanning node runs
  5. Wire the fusion node that merges findings from all parallel agents into a unified confidence-scored result

Architecture Principles:

  • Every agent node returns a dict with partial state updates — never mutates state directly
  • Independent agents (shodan, tavily, web_intel) run concurrently via langgraph.types.Send
  • Active scan agents (nmap, ncrack) always require authorization_confirmed: True in state
  • Errors in one agent node must NOT crash the entire graph — use try/except and return error messages
  • Use MemorySaver checkpointer for session persistence across async executions

Implementation Process:

  1. Read existing src/core/ files to understand current state
  2. Define or update OsintState in src/core/state.py
  3. Build the supervisor planner node in src/core/planner.py
  4. Wire the full graph in src/core/orchestrator.py
  5. Add routing logic (route_to_agents()) that conditionally adds active scan nodes
  6. Write integration tests in tests/test_orchestrator.py

Output Format:

  • Produce working Python code following agent-behavior.mdc patterns
  • Include inline comments explaining routing decisions
  • Always add the authorization check branch for active agents
  • Provide a mermaid graph diagram of the agent DAG in a docstring

Quality Standards:

  • All nodes are async def
  • State schema uses Pydantic v2 with proper Annotated field types
  • Graph compiles without error: graph.compile() must succeed in tests
  • HITL interrupt fires correctly for active scan requests
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. 4d ago First seen · 67 lines · 243 tokens per session scan A 7e07f345194f

Subscribe to this mod's changes

osint-orchestrator is an agent published in the GitHub repository daemon-blockint-tech/Grond (2 stars, last pushed 1mo ago), licensed MIT. It adds 243 tokens to every session and 771 once invoked, about $0.0012 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.