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 instructions/timothyvang/verdict-dfir/agents-mdgit clone --depth 1 https://github.com/TimothyVang/verdict-dfirWrote 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/instructions/timothyvang/verdict-dfir/agents-md)<a href="https://agentmods.dev/instructions/timothyvang/verdict-dfir/agents-md"><img src="https://agentmods.dev/badge/instructions/timothyvang/verdict-dfir/agents-md.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 5.1 | $0.03098 | $0.03098 |
| Opus 5 | $0.01549 | $0.01549 |
| Sonnet 5 | $0.00620 | $0.00620 |
| Haiku 4.5 | $0.00310 | $0.00310 |
Grade A, and why
verdict-dfir AGENTS.md 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 today.
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 — 170 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Agent instructions for VERDICT DFIR. This file is for Codex, OpenCode, and other coding agents that follow the AGENTS.md convention. Claude Code also reads CLAUDE.md; for Claude-specific runtime behavior, CLAUDE.md is authoritative.
VERDICT ecosystem — what is what
VERDICT is a local-first DFIR (digital forensics & incident response) agent platform, split into three repos:
| Repo | Role | It is… |
|---|---|---|
caseforge-core |
Headless controller: privacy routing, model selection, structured findings, custody validation, the caseforge CLI. |
the driver |
verdict-opencode |
The agent runtime — a branded fork of opencode; the verdict binary is built from it. |
the engine |
| verdict-dfir-beta / this repo | The forensic toolkit: findevil-mcp (Rust, 32 read-only tools) + findevil-agent-mcp (Python, 14 custody/crypto tools) + DFIR doctrine (agent-config/) + hash-chained custody. Consumed by caseforge via VERDICT_DFIR_HOME. |
the evidence lab (you are here) |
Runtime flow: caseforge (controls + guards) → verdict binary (runs the agent) → this repo's findevil MCP tools (do the forensics) → hash-chained custody → caseforge verify.
Two rules everything obeys: the LLM is not the forensic source of truth (findings must cite a tool_call_id + output_sha256 + verified manifest); real evidence stays local by default.
Start Here
- Work from the repository root.
- Install or verify prerequisites with
bash scripts/setup; usebash scripts/doctor.shfor a preflight summary. - The canonical product run is
scripts/verdict <evidence>. - In Claude Code, the equivalent operator shortcut is
/verdict <evidence>orinvestigate <path>. - Do not create or revive a separate Product CLI.
scripts/verdict,scripts/find-evil, and Claude Code are the supported entry points. - Before changing investigation behavior, read
CLAUDE.mdand the runtime files inagent-config/.
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.
- today First seen · 170 lines · 3,098 tokens per session scan A 0a13617118f1
verdict-dfir AGENTS.md is an instructions file published in the GitHub repository TimothyVang/verdict-dfir (12 stars, last pushed 2d ago), licensed Apache-2.0. It adds 3,098 tokens to every session, about $0.0155 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 2026-09-04.
Other instructions, from other repositories
SAVVYDFIR-MCP CLAUDE.md
Instructions for kismatkunwar89/SAVVYDFIR-MCP, covering savvydfir-mcp - dfir investigation framework, what this is, critical rules, forensic knowledge in tool responses and investigation workflow (7 phases).
sift-mcp AGENTS.md
Instructions for AppliedIR/sift-mcp, covering valhuntir — forensic investigation platform, getting started, available mcp servers, malware analysis escalation and investigation recording.
ai-dfir-toolkit CLAUDE.md
Instructions for depalmar/ai-dfir-toolkit, covering project context, what this is, commands, rules that are not negotiable and what a restricted runner cannot verify.
mem_forensics-mcp CLAUDE.md
Claude Code instructions for x746b/mem_forensics-mcp, covering role: memory forensics specialist, mem-forensics-mcp: triage first, architecture, investigation questions -> tools and workflow.
steganography-mcp CLAUDE.md
Claude Code instructions for badchars/steganography-mcp, covering steganography-mcp — steganography analysis mcp server, architecture, key rules, categories (17) and commands.
agentic-dart CLAUDE.md
Instructions for Juwon1405/agentic-dart, covering claude.md — ai assistant guide for agentic-dart, what this project is, repository map, run modes (know which one you are touching) and non-negotiables (do not break these).