attack-selection-guide SKILL.md skillA
attack-selection-guide from dreadnode/capabilities is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add dreadnode/capabilities/attack-selection-guide@git:20260716.236a798Pins git:20260716.236a798 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- dreadnode/capabilities
- Path
capabilities/ai-red-teaming/skills/attack-selection-guide/SKILL.md- Versions recorded
- 2
- Latest version
- git:20260716.236a798, committed Jul 16, 2026
- First version recorded
- May 4, 2026
- Size
- 13,737 bytes
- Repository stars
- 21
- License
- MIT
Decision tree for selecting AIRT attacks based on goals, target type, and constraints.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Attack Selection Guide
Use this decision tree to select the right attacks for an assessment.
## Step 1: Identify Target Type
### LLM / Chat Model
Any provider (OpenAI, Anthropic, Google, Groq, Mistral, Together, AWS Bedrock, Azure, Ollama, custom).
Use `generate_attack` with `target_model="provider/model"`. All jailbreak attacks apply.
### Agentic System (Tool-Using Agent)
Multi-step agents that call tools, access MCP servers, or interact with other agents.
Use `generate_agentic_attack` with the appropriate preset (`openai_assistants`, `anthropic`, `custom`).
### Custom API Endpoint
Any HTTP endpoint. Use `generate_agentic_attack` with `agent_preset="custom"` and provide request/response templates.
### Multi-Agent System (Delegation Chains, Trust Boundaries)
A system of cooperating agents (entry → mid → privileged) where the attack must
cross trust boundaries or propagate through delegation to make a downstream agent
execute a tool. Use `generate_atlas_attack` (ATLAS — Adaptive Topology-Level Attack
Synthesis). Point it at a deployed environment exposing
`POST /attack {prompt, surface, injection} -> {content, tool_calls, ...}`.
…Read the whole file at its exact version.
How to install
mdr add dreadnode/capabilities/attack-selection-guide@sha256:48b57f5b0c25dd24The line at the top of this page pins the label git:20260716.236a798, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_xsuolk2eeckofgg7)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 2 versions of this file, between May 4, 2026 and Jul 16, 2026, about one every 72 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260716.236a798 latest | 2026-07-16 | 236a798 | 13,737 B | A | view · diff |
| git:20260504.2a64374 | 2026-05-04 | 2a64374 | 12,834 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (13737 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
dreadnode/capabilities · 21 stars · license MIT · pushed 2026-11-02 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_xsuolk2eeckofgg7 GET https://markdownregistry.com/api/v1/resolve?ref=dreadnode/capabilities/attack-selection-guide GET https://markdownregistry.com/api/v1/blob/48b57f5b0c25dd24de60fd185d36be388a66487f859c98073a3142d5836bb859
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.