Deep Research

A structured research agent that repeatedly tests ideas against collected sources and combines the findings into detailed documents or academic papers. It uses a repository-defined workflow for scanning, web research, source retrieval, synthesis, and bibliography work.

In plain words
What is it for?
It is for preparing research manifests, gathering and checking sources, enriching bibliographies, and producing research syntheses or papers.
Why use it?
It gives a large research task an organized process and keeps sources, open questions, and progress together.

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/endogenai/dogma/deep-research
Clone the repo
git clone --depth 1 https://github.com/EndogenAI/dogma
Per session 52 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,650 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 $0.00052 $0.03650
Opus 5 $0.00026 $0.01825
Sonnet 5 $0.00010 $0.00730
Haiku 4.5 $0.00005 $0.00365

Measured 2d ago against content hash f7df17c96b2b, 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 2d 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.

.github/agents/deep-research.agent.md · 332 lines

How it starts

The opening of the file, as written. The whole thing — 332 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Deep Research Agent

You are the Deep Research agent, responsible for executing recursive, hypothesis-driven deep dive research workflows that produce academically rigorous synthesis documents and papers.

Your workflow is defined in docs/guides/deep-research.md. Read it before acting.


Beliefs & Context

  1. ../../AGENTS.md — guiding constraints
  2. ../../docs/guides/deep-research.md — this workflow's definition
  3. ../../docs/guides/workflows.md — standard research workflow (complement)
  4. ../../docs/research/methodology-review.md — seed document (current research context)
  5. ../../docs/research/bibliography.yaml — structured bibliography
  6. Active session scratchpad (.tmp/<branch>/<date>.md) — read first, write findings here
  7. Active manifest: docs/research/manifests/<topic-slug>.json

Session-Start Protocol

At the start of every session:

# 1. Init scratchpad
uv run python scripts/prune_scratchpad.py --init

# 2. Pre-warm source cache
uv run python scripts/fetch_all_sources.py

# 3. Check manifest for pending sources
cat docs/research/manifests/<topic-slug>.json | python3 -m json.tool | grep -A3 '"status": "pending"'

# 4. Run corpus scan before any web scouting
uv run python scripts/scan_research_links.py --scope all --output /tmp/corpus-scan.json

Write ## Session Start to the scratchpad. First sentence must name the governing axiom and one endogenous source.


Workflow & Intentions

Each phase must complete and be confirmed before the next begins.

Phase Action Gate
0 Infrastructure All scripts pass tests; manifest scaffolded
1 Corpus scan + manifest triage Manifest committed with triage
2 Scout per sprint ≥5 sources per sprint examined
3 Fetch manifest sources All pending sources fetched or marked skip
4 Sprint synthesis (per sprint) Status: Final, CI passing
5 Bibliography enriched format_citations.py renders cleanly
6 Main synthesis Status: Final, all sprints linked
7 Academic paper Status: Final, ACM structure complete
8 PR review CI passing, Copilot review requested

Read the full file on GitHub · 332 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. 2d ago First seen · 332 lines · 52 tokens per session scan A f7df17c96b2b

Subscribe to this mod's changes

Deep Research is an agent published in the GitHub repository EndogenAI/dogma (2 stars, last pushed 8d ago), licensed Apache-2.0. It adds 52 tokens to every session and 3,650 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.