agentleFS
Sign inStart free
kaandgnshkaandgnsh/Claude-Blue/skills/ai-security/llm-gateway-controls/SKILL.md

llm-gateway-controls

Skill0 starsChanged 13 days agoMITView on GitHub

Design and review controls at an LLM/AI gateway or proxy (LiteLLM, cloud AI gateways, Azure API Management AI policies, commercial AI firewalls) — authentication and per-user attribution, approved models and providers, prompt/response logging with privacy, DLP and secret redaction, injection/jailbreak classifiers, rate and spend limits, and shadow-AI discovery. Use when governing enterprise use of LLM APIs, preventing sensitive data leakage to AI services, or monitoring AI usage.

  • Reads credentials

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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
  1. Distinctive:

    Check the current LiteLLM docs for guardrail and per-key budget configuration syntax.

    line 46

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

LLM Gateway Controls

Control set

ControlPurpose
Central gateway for all LLM API trafficVisibility, single enforcement point; block direct egress to AI provider APIs from servers
Strong auth + per-user/app identityAttribution of prompts and spend; no shared provider keys in apps
Model/provider allow-listOnly approved models/regions/data-processing terms (no-training, retention terms)
Input DLP & secret redactionDetect and mask secrets, credentials, PII, classified content before sending (telemetry-minimization patterns)
Injection/jailbreak detectionClassifiers on inputs and retrieved content (e.g., prompt shields); log and optionally block
Output filteringPII, secrets, unsafe content; markdown/image URL stripping for exfil prevention
Rate limits & budgetsPer user/app token and cost caps (LLM10)
LoggingPrompts/responses/tool calls to restricted store with retention and access audit; metadata to SIEM
Key managementProvider keys in a vault, rotated; alert on usage anomalies (LLMjacking)

Shadow AI discovery

Proxy/DNS logs for AI SaaS domains (chat assistants, transcription, code assistants), CASB app catalogs, browser extension inventory, OAuth grants to AI apps (oauth-app-review), expense reports. Provide an approved alternative rather than only blocking.

Example: LiteLLM proxy config fragment (illustrative)

model_list:
  - model_name: approved-chat
    litellm_params:
      model: azure/<deployment>
      api_base: os.environ/AZURE_API_BASE
      api_key: os.environ/AZURE_API_KEY
general_settings:
  master_key: os.environ/LITELLM_MASTER_KEY
litellm_settings:
  max_budget: 1000
  budget_duration: 30d

Check the current LiteLLM docs for guardrail and per-key budget configuration syntax.

Review output

Control | Implemented | Coverage (apps/users) | Evidence | Gap | Recommendation

Related skills

ai-incident-response, prompt-injection-review, dlp-alert-triage, shadow-it-discovery

References

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free