dma-attack-techniques

Guide for PCIe DMA threat modeling, FPGA-based memory access, and defensive implications in game security. Use this skill when researching pcileech, BAR and TLP behavior, page-table walking, IOMMU or VT-d, device impersonation, firmware mimicry, or DMA detection and mitigation in game security research.

Quality

63%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/dma-attack/SKILL.md

Quality

Content

27%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is an encyclopedic reference dump — verbose, largely conceptual, and monolithic with broken wiki references — containing only scattered executable snippets and partial workflows rather than a lean, actionable, well-split skill. The description is strong, but the content wastes the context window on knowledge Claude already has.

Suggestions

Replace the ~1760-line reference exposition with a concise overview and split the PCIe/IOMMU/FPGA/pcileech details into separate reference files under references/, linking one level deep instead of inlining everything in SKILL.md.

Remove or relocate concepts Claude already knows (PCIe TLP format, CAM/ECAM, MSI/MSI-X basics, PCR allocation) and keep only the domain-specific, non-obvious guidance and executable commands.

Verify the referenced wiki paths (wiki/index.md, wiki/AGENTS.md, wiki/overviews/dma-attack.md) actually exist as bundle files, or drop the dangling references; consolidate the executable commands and snippets into a single actionable quick-start section.

Dimension	Reasoning	Score
Conciseness	The ~1760-line body is dense reference exposition — PCIe protocol tables, capability offsets, IOMMU/VT-d internals, PCR allocations — largely concepts Claude already knows and padded with explanatory prose, matching the verbose/explains-known-concepts anchor rather than the mostly-efficient anchor at 2.	1 / 3
Actionability	Scattered concrete items exist (a pcileech_read_phys snippet, a TranslateVA C function, `lspci -d [VID:DID] -vvv -xxxx`, `setpci -s <BDF> ECAP_AER+0x4.L`, archive URL formats), but they are sparse islands in mostly descriptive prose and not copy-paste ready as a whole, fitting the some-concrete-guidance-but-incomplete anchor.	2 / 3
Workflow Clarity	Some sequenced pipelines with checkpoints appear (Layered Detection Pipeline, Multi-Signal Correlation, "containment protects the live match"), but large sections are flat catalogs (16-row bypass catalog, firmware tier table) with no sequence or validation, so checkpoints are present in places and missing in others.	2 / 3
Progressive Disclosure	It is a single monolithic SKILL.md with no bundle files (references/scripts/assets absent) and inline reference content that should be split out; the referenced `wiki/index.md`, `wiki/AGENTS.md`, and `wiki/overviews/dma-attack.md` paths do not exist as actual bundle files, matching the monolithic-wall-of-text anchor rather than the some-structure anchor at 2.	1 / 3
	Total	6 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, trigger-rich, and complete: it states concrete capabilities, an explicit "Use this skill when..." trigger covering natural domain terms, and a distinctive niche that is unlikely to conflict with other skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete capabilities — "PCIe DMA threat modeling, FPGA-based memory access, and defensive implications" plus researching named subsystems (pcileech, BAR/TLP, page-table walking, IOMMU/VT-d, device impersonation, firmware mimicry) — matching the anchor for listing multiple specific concrete actions, not just naming a domain.	3 / 3
Completeness	Explicitly answers both what ("Guide for PCIe DMA threat modeling, FPGA-based memory access, and defensive implications in game security") and when with an explicit "Use this skill when researching..." trigger clause, satisfying the anchor for clearly answering both what AND when.	3 / 3
Trigger Term Quality	Covers natural terms a researcher in this domain would actually say — "pcileech, BAR and TLP behavior, page-table walking, IOMMU or VT-d, device impersonation, firmware mimicry, or DMA detection and mitigation in game security research" — giving broad coverage rather than a thin keyword set.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche — PCIe DMA / pcileech / IOMMU-VT-d research in game security — with distinct, specialized triggers unlikely to fire for unrelated skills, matching the anchor for a clear niche with distinct triggers.	3 / 3
	Total	12 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1759 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: gmh5225/awesome-game-security
Path: .claude/skills/dma-attack/SKILL.md
Commit: 9efdfd0

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.