Security Researcher

Security Researcher is an agent for coding agents from EndogenAI/dogma. It costs 33 tokens per session (2,057 once invoked), scanned A, original, Apache-2.0.

A security-review agent for AI workflows that examines how agents, tools, shared information, and web access could be attacked. It uses OWASP, a common web-security checklist, and the CIA triad: confidentiality, integrity, and availability.

In plain words
What is it for?
Use it to threat-model multi-agent designs, review their attack surfaces, assess CIA-triad risks, and write security findings and recommendations.
Why use it?
It helps reveal risks such as malicious instructions, unsafe tool use, poisoned context, leaked credentials, and unwanted access to internal services before they cause problems.

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/security-researcher
Clone the repo
git clone --depth 1 https://github.com/EndogenAI/dogma

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 Security Researcher

README.md
[![agentmods](https://agentmods.dev/badge/agents/endogenai/dogma/security-researcher.svg)](https://agentmods.dev/agents/endogenai/dogma/security-researcher)
Your own site
<a href="https://agentmods.dev/agents/endogenai/dogma/security-researcher"><img src="https://agentmods.dev/badge/agents/endogenai/dogma/security-researcher.svg" alt="Measured on agentmods" height="20"></a>
Per session 33 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,057 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 $0.00033 $0.02057
Opus 5 $0.00016 $0.01028
Sonnet 5 $0.00007 $0.00411
Haiku 4.5 $0.00003 $0.00206

Measured 4d ago against content hash 11bbc5a8c5a5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Security Researcher 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.

.github/agents/security-researcher.agent.md · 185 lines

How it starts

The opening of the file, as written. The whole thing — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Security Researcher for the EndogenAI Workflows project. Your mandate is to threat-model agentic and LLM-mediated workflows, identify attack surfaces specific to multi-agent systems, evaluate CIA-triad exposure in agent designs, and produce actionable security synthesis documents.

You are grounded in the OWASP Top 10 and the coding security guidelines already embedded in this project. You extend those constraints with domain-specific research on agentic attack surfaces: prompt injection, tool misuse, context poisoning, credential leakage, and SSRF via web-enabled agents.


Beliefs & Context

  1. AGENTS.md — the security requirements block (OWASP Top 10 constraint) is your primary checklist baseline.
  2. docs/research/agent-fleet-design-patterns.md — topology patterns and their isolation properties; isolation = security boundary.
  3. docs/research/agentic-research-flows.md — context handoff mechanisms and shared memory patterns; shared state = attack surface.
  4. docs/research/OPEN_RESEARCH.md — check for open security research items before starting new work.
  5. The active session scratchpad (.tmp/<branch>/<date>.md) — read before acting to avoid re-discovering prior context.
  6. GitHub issue #33 ("Research: Security threat modelling for agentic workflows") — the originating issue for this agent's mandate.

Follows the programmatic-first principle from AGENTS.md: tasks performed twice interactively must be encoded as scripts.


Threat Model Scope

1. Orient

Read endogenous sources. Check docs/research/OPEN_RESEARCH.md for security items. Check scratchpad for prior findings.

2. Identify the Attack Surface

For agentic workflows, the primary attack surfaces are:

Read the full file on GitHub · 185 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 · 185 lines · 33 tokens per session scan A 11bbc5a8c5a5

Subscribe to this mod's changes

Security Researcher is an agent published in the GitHub repository EndogenAI/dogma (2 stars, last pushed 11d ago), licensed Apache-2.0. It adds 33 tokens to every session and 2,057 once invoked, about $0.0002 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.