rag-security-review
Skill0 starsChanged 13 days agoMITView on GitHub
Review retrieval-augmented generation (RAG) systems — document-level access control at retrieval time, ingestion pipeline trust (poisoned or injected documents), vector store security and tenant isolation, embedding inversion/leakage, citation integrity, and logging. Use when building or assessing enterprise search copilots, chat-with-your-docs features, or vector databases (OWASP LLM08:2025).
1 other project uses a version of this.
Adopted by
Most recently taken up by
- fevziegeyurtsevenler/llm-security-skillstheir own version, first seen Jul 25, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- fevziegeyurtsevenler/llm-security-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository273 other files it gives its agents.
- ad-attack-path-analysisSkill
- adcs-security-reviewSkill
- ad-privileged-access-reviewSkill
- dcsync-detectionSkill
- gpo-security-reviewSkill
- kerberos-attack-detectionSkill
- agent-tool-permission-reviewSkill
- ai-incident-responseSkill
- llm-gateway-controlsSkill
- llm-security-evaluationSkill
- mcp-server-security-reviewSkill
- model-artifact-securitySkill
- prompt-injection-reviewSkill
- api-abuse-detectionSkill
- api-authorization-testingSkill
- api-inventorySkill
- api-schema-securitySkill
- dast-baselineSkill
- sast-triageSkill
- sca-dependency-riskSkill
- secure-code-reviewSkill
- security-requirementsSkill
- threat-modelingSkill
- asset-criticality-tieringSkill
- asset-inventory-reconciliationSkill
- external-attack-surface-reviewSkill
- shadow-it-discoverySkill
- aws-credential-compromiseSkill
- aws-iam-reviewSkill
- aws-org-guardrailsSkill
- cloudtrail-investigationSkill
- guardduty-triageSkill
- s3-exposure-reviewSkill
- azure-activity-investigationSkill
- azure-rbac-reviewSkill
- conditional-access-reviewSkill
- entra-audit-investigationSkill
- entra-signin-investigationSkill
- backup-assuranceSkill
- critical-service-mappingSkill
- ransomware-recoverySkill
- recovery-exerciseSkill
- cloud-exposure-reviewSkill
- cloud-incident-responseSkill
- cloud-logging-baselineSkill
- cloud-posture-assessmentSkill
- audit-evidence-packageSkill
- continuous-control-monitoringSkill
- control-mappingSkill
- control-testing-samplingSkill
- container-incidentSkill
- dockerfile-hardeningSkill
- image-scanningSkill
- runtime-detection-falcoSkill
- db-audit-loggingSkill
- db-exfiltration-investigationSkill
- db-hardeningSkill
- db-privilege-reviewSkill
- data-classificationSkill
- data-discoverySkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 30
Test (authorized test users): user without access to doc X asks questions only X can answer — expect refusal/no citation.
- Distinctive: line 31
Check vector DB security: authentication, network exposure, encryption, admin access, backups.
- Distinctive: line 32
Verify citations link to source with the user's own access.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
RAG Security Review
Key risks
| Risk | Description | Control |
| Permission bypass | Retriever returns chunks the user can't access in the source system | Enforce source ACLs at query time (filter by user/group IDs synced from source; re-check on read), not only at ingestion |
| Over-sharing amplification | RAG surfaces content that was technically shared too broadly ("everyone" links) | Fix source permissions; sensitivity-label-aware retrieval; exclude restricted labels |
| Indirect prompt injection | Malicious instructions inside indexed docs | prompt-injection-review; restrict who can write to indexed sources; content provenance in prompts |
| Data poisoning | Attacker inserts false info to steer answers | Source allow-lists, ingestion review for external sources, provenance and citations |
| Cross-tenant leakage | Shared vector index without tenant filter | Separate indexes/namespaces per tenant or mandatory tenant filter enforced server-side |
| Embedding leakage | Embeddings can be partially inverted to recover text | Treat vectors as sensitive as the source text; encrypt, restrict access |
| Stale/deleted data | Deleted source documents remain in the index | Deletion propagation, re-index schedules, retention |
| Logging exposure | Prompts/answers containing sensitive data stored widely | Restricted logs, retention, redaction (telemetry-minimization) |
Review steps
- Diagram ingestion → chunking → embedding → vector store → retrieval → prompt assembly → LLM → output.
- For each source: who can write, how ACLs are captured, sync frequency, deletion handling.
- Test (authorized test users): user without access to doc X asks questions only X can answer — expect refusal/no citation.
- Check vector DB security: authentication, network exposure, encryption, admin access, backups.
- Verify citations link to source with the user's own access.
Output
Source | Write access | ACL enforcement point | Tenant isolation | Injection exposure | Findings | Fix
Related skills
prompt-injection-review, data-classification, llm-gateway-controls, trust-boundary-analysis
References
- OWASP LLM08:2025 Vector and Embedding Weaknesses: https://genai.owasp.org/llmrisk/llm082025-vector-and-embedding-weaknesses/
- OWASP LLM02:2025 Sensitive Information Disclosure: https://genai.owasp.org/llmrisk/llm022025-sensitive-information-disclosure/
- NIST AI 600-1 Generative AI Profile: https://www.nist.gov/publications/artificial-intelligence-risk-management-framework-generative-artificial-intelligence
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.