SecureAEye LAB ONLINE · SIG 657 · DET ~41s

← All attacks / AI/ML and emerging attacks

AI coding-assistant config / key theft

Infostealers target .claude/.cursor dirs and config files for LLM keys

categoryAI/ML and emerging attacksseverityMediumdetectabilityHighlabnot yet measured

Example: ACRSTEALER, VIDAR

Classification

30 values: 18 imported, 2 derived, 10 machine-classified. What do these mean?

Exploitability

Attack vector	AI model manipulationCredential attacks
Reach (CVSS)	Local
Attack complexity	Low
Privileges required	Low
User interaction	Active
Attacker skill required	Low
Resources required	Low
Likelihood / prevalence	High
Maturity	Proof of concept

Impact & scope

Severity	Medium
Impact	ConfidentialityFinancial
Platform	AI SystemContainersIaaSLinuxmacOSWindows
Environment	AI system
Target asset class	ML model / agent
Kill-chain phase (tactic)	Credential Access

Detection & telemetry

Detectability	High
Testability	Atomic Red Team test
Data sources (ATT&CK data components)	Command ExecutionFile AccessFile CreationLogon Session CreationNetwork Connection CreationProcess Creation

Detection & response

Mitigated by 24

M1017 attack-mitigation
M1022 attack-mitigation
M1027 attack-mitigation
M1047 attack-mitigation
D3-ANCI d3fend
D3-CCSA d3fend
D3-CF d3fend
D3-CH d3fend
D3-CM d3fend
D3-CQ d3fend
D3-CR d3fend
D3-CRO d3fend
D3-CTS d3fend
D3-DF d3fend
D3-DUC d3fend
D3-FA d3fend
D3-FE d3fend
D3-FEV d3fend
D3-FIM d3fend
D3-LFP d3fend
D3-MFA d3fend
D3-RF d3fend
D3-RFAM d3fend
D3-RIC d3fend

Detected by 25

DET0307 attack-detection-strategy
04936b66-3915-43ad-a8e5-809eadfd1141 sigma
0aba5685-6db6-486f-88ef-29a99c545cfd sigma
0cf7a157-8879-41a2-8f55-388dd23746b7 sigma
1a0f6f16-2099-4753-9a02-43b6ac7a1fa5 sigma
3b2c1059-ae5f-40b6-b5d4-6106d3ac20fe sigma
44030449-b0df-4c94-aae1-502359ab28ee sigma
459a2970-bb84-4e6a-a32e-ff0fbd99448d sigma
53b1b378-9b06-4992-b972-dde6e423d2b4 sigma
587254ee-a24b-4335-b3cd-065c0f1f4baa sigma
6d3a3952-6530-44a3-8554-cf17c116c615 sigma
6ea858a8-ba71-4a12-b2cc-5d83312404c7 sigma
76a26006-0942-430b-8249-bd51d448f8e5 sigma
80eeab92-0979-4152-942d-96749e11df40 sigma
851fd622-b675-4d26-b803-14bc7baa517a sigma
b831353c-1971-477b-abb6-2828edc3bca1 sigma
b83f5166-9237-4b5e-9cd4-7b5d52f4d8ee sigma
bd5971a7-626d-46ab-8176-ed643f694f68 sigma
cd072b25-a418-4f98-8ebc-5093fb38fe1a sigma
d557dc06-62e8-4468-a8e8-7984124908ce sigma
d7a650c4-226c-451e-948f-cc490db506aa sigma
df3fcaea-2715-4214-99c5-0056ea59eb35 sigma
f4ff7323-b5fc-4323-8b52-6b9408e15788 sigma
f576a613-2392-4067-9d1a-9345fb58d8d1 sigma
fa4aaed5-4fe0-498d-bbc0-08e3346387ba sigma

Tested by 32

00e3e3c7-6c3c-455e-bd4b-461c7f0e7797 atomic
0d4f2281-f720-4572-adc8-d5bb1618affe atomic
0e56bf29-ff49-4ea5-9af4-3b81283fd513 atomic
114dd4e3-8d1c-4ea7-bb8d-8d8f6aca21f0 atomic
14575d1f-16b3-4a0b-8986-540e42e6a6cc atomic
18318786-e0c0-4bdf-979f-f022e6f5f2ef atomic
2951c09f-b4f9-4248-ba75-6373709ebd55 atomic
2a24627b-061b-43e6-bece-a5b0638841dd atomic
367d4004-5fc0-446d-823f-960c74ae52c3 atomic
37807632-d3da-442e-8c2e-00f44928ff8f atomic
51cde30c-0311-449a-8e15-542b59f95939 atomic
5312c7c3-775b-4437-a8b0-914b30cfebd4 atomic
5a668739-3a2f-4265-895f-c0e1e0b09eaa atomic
75f66e03-37d3-4704-9520-3210efbe33ce atomic
98618067-a2c7-4d4b-9c89-403a90f43656 atomic
9d9c22c9-fa97-4008-a204-478cf68c40af atomic
9e507bb8-1d30-4e3b-a49b-cb5727d7ea79 atomic
a11f22fb-a037-4cc2-bcfc-a133b287aa1b atomic
a8f6148d-478a-4f43-bc62-5efee9f931a4 atomic
aa12eb29-2dbb-414e-8b20-33d34af93543 atomic
aaa87b0e-5232-4649-ae5c-f1724a4b2798 atomic
af664fc0-0772-4866-a661-d49a22a13487 atomic
b0cdacf6-8949-4ffe-9274-a9643a788e55 atomic
bcf8e58a-1cb7-4988-9227-3d3c70751811 atomic
bd4cf0d1-7646-474e-8610-78ccf5a097c4 atomic
c3372ae9-b9bf-427a-b330-cb04fecb9c1d atomic
c8367142-cbb0-47a4-96c7-ff9969fd64ea atomic
c9dc9de3-f961-4284-bd2d-f959c9f9fda5 atomic
d1665562-c389-49fc-9cbe-92236816a105 atomic
da4f751a-020b-40d7-b9ff-d433b7799803 atomic
e62bccc4-0200-47e9-b40a-4ea5328495b3 atomic
fdd0c913-714b-4c13-b40f-1824d6c015f2 atomic

Attack relationships

Parent: API key / secret leakage
Parent: CI/CD credential hygiene failures

Framework references

Primary source: https://cloud.google.com/blog/topics/threat-intelligence/from-prompting-to-autonomy-the-evolution-of-adversarial-ai/
atlas: AML.T0083 (primary)
attack: T1552.001

Evidence & sources

ev-0001 (Cyber Attack Taxonomy) — Name, definition and Example/ID taken verbatim from the 'AI/ML and emerging attacks' table, line 640 of Cyber-Attack-Taxonomy.md
ev-0002 (Seeded-record enrichment run 2026-09-30 (crosswalk + facet classification)) — Crosswalk to atlas:AML.T0083, attack:T1552.001 as the nearest framework entries; ids verified to exist in the cached catalogues.
ev-0003 (MITRE Enterprise ATT&CK v19.2 (STIX 2.1)) — ATT&CK usage prevalence: 41 groups/software/campaigns reference this technique (ATT&CK enterprise v19.2)