deep-research

deep-research is a command for Codex from marcusquinn/aidevops. It costs 13 tokens per session (1,119 once invoked), scanned A, original, MIT.

A command for producing a documented research report with cited sources, supporting evidence, and related planning files. It can take a topic or a written brief and save the results as a reusable bundle.

In plain words
What is it for?
Use it for topics such as comparing hosted databases or assessing EU AI Act obligations. It can create a report, source ledger, claim ledger, outline, and evidence files.
Why use it?
It removes the work of organizing research into a consistent record of sources, claims, questions, and raw evidence. This makes the reasoning behind a decision easier to review.

Command for Codex

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 commands/marcusquinn/aidevops/deep-research
Clone the repo
git clone --depth 1 https://github.com/marcusquinn/aidevops

Made for: Codex.

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 deep-research

README.md
[![agentmods](https://agentmods.dev/badge/commands/marcusquinn/aidevops/deep-research.svg)](https://agentmods.dev/commands/marcusquinn/aidevops/deep-research)
Your own site
<a href="https://agentmods.dev/commands/marcusquinn/aidevops/deep-research"><img src="https://agentmods.dev/badge/commands/marcusquinn/aidevops/deep-research.svg" alt="Measured on agentmods" height="20"></a>
Per session 13 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,119 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 $0.00013 $0.01119
Opus 5 $0.00006 $0.00560
Sonnet 5 $0.00003 $0.00224
Haiku 4.5 $0.00001 $0.00112

Measured 4d ago against content hash 9a5e19571ea8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

deep-research 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/scripts/commands/deep-research.md · 142 lines

How it starts

The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Produce a cited, decision-grade research deliverable.

Arguments: $ARGUMENTS

Use /auto-research instead when the task is an autonomous experiment loop that modifies files, measures a metric, and keeps or discards changes. Use /deep-research when the output is a durable artifact: cited report, source ledger, claim ledger, outline, and raw evidence bundle.

Invocation Patterns

Pattern Example Behaviour
Topic /deep-research "compare hosted vector databases for RAG" Infer scope and produce a report
Brief file /deep-research --brief todo/research/vector-db-brief.md Read scope, questions, and constraints from file
Output dir /deep-research "EU AI Act vendor obligations" --out todo/research/eu-ai-act Write deliverables to the requested directory
Headless /deep-research --brief path/to/brief.md --out path/to/out Run without questions using the supplied scope

Output Contract

Create an export bundle under todo/research/deep-{slug}/ unless --out is provided:

todo/research/deep-{slug}/
├── source-plan.md       # intended source classes and search strategy
├── sources.tsv          # source ledger with URL/title/publisher/date/accessed/use
├── claims.tsv           # claim ledger mapping claims to source IDs and confidence
├── outline.md           # report structure before final synthesis
├── report.md            # cited final report
└── raw/                 # saved excerpts or fetched artifacts when safe and permitted

Workflow

Step 1: Scope

Parse topic, brief path, output path, deadline, geography, audience, and decision to support. If no brief is supplied, infer a concise scope from $ARGUMENTS and record it in source-plan.md. In headless mode, do not ask questions; document assumptions in the plan.

Step 2: Source Plan

Write source-plan.md before collecting evidence:

Read the full file on GitHub · 142 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. 4d ago First seen · 142 lines · 13 tokens per session scan A 9a5e19571ea8

Subscribe to this mod's changes

deep-research is a command published in the GitHub repository marcusquinn/aidevops (391 stars, last pushed today), licensed MIT. It adds 13 tokens to every session and 1,119 once invoked, about $0.0001 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 2026-08-30.