Developer Packages
Open-source packages for AI security, inventory instrumentation, and evidence generation. Bounded components within HAIEC evidence-bound assurance.
Packages are bounded components, not the full HAIEC product.
These packages provide specific security checks, instrumentation, or evidence generation. They do not by themselves constitute HAIEC Evidence-Bound Assurance, certification, or a compliance guarantee. Learn what HAIEC is.
Developer Security Packages
Three independent security checks: source code, tenant boundaries, and model I/O. One does not perform the functions of another.
AI AppSec
Evidence-backed static security analysis for AI applications and agents.
npm install -g ai-appsecLimitations:
- •Static analysis only — no runtime behavioral evidence
- •No compliance certification or assessment
- •No tenant isolation checking (roadmap)
MCP Tenant Isolation
Static analysis scanner for MCP server code and multi-tenant SaaS. Detects cross-tenant data leakage before production.
npm install -g mcp-tenant-isolation57 rules
Limitations:
- •Intra-procedural flow analysis only — does not trace across function calls, files, or middleware boundaries
- •No runtime verification — cannot verify database-level RLS enforcement or middleware execution
- •TypeScript and JavaScript only — Python, Go, Ruby not supported yet
LLMVerify
Local-first verification, PII redaction, prompt-injection defense, runtime monitoring, and a built-in MCP server for any LLM.
npm install llmverifyLimitations:
- •Cannot definitively prove hallucinations — signals are pattern-based, not ground-truth verification
- •Does not replace human review — use to triage, not to approve
- •PII detection is regex-based (~90% for standard formats, lower for variations)
Inventory Instrumentation
Drop-in SDK wrappers for AI Inventory — not security scanners
@haiec/openai
OpenAI SDK wrapper with automatic usage tracking and cost calculation for HAIEC AI Inventory.
npm install @haiec/openaiLimitations:
- •Instrumentation only — does not scan code or verify security
- •Not a security scanner — does not detect vulnerabilities
- •Requires HAIEC AI Inventory for data aggregation and reporting
@haiec/anthropic
Anthropic Claude SDK wrapper with automatic usage tracking and cost calculation for HAIEC AI Inventory.
npm install @haiec/anthropicLimitations:
- •Instrumentation only — does not scan code or verify security
- •Not a security scanner — does not detect vulnerabilities
- •Requires HAIEC AI Inventory for data aggregation and reporting
Not security scanners. These packages instrument usage into HAIEC AI Inventory. For security checks, see Developer Security.
Evidence Generation
Python libraries for AI training lineage evidence
ISAF Logger
Instruction Stack Audit Framework — automatic lineage logging for AI training pipelines.
pip install isaf-loggerLimitations:
- •Evidence generation utility — not a security scanner
- •Does not validate model behavior, runtime safety, or regulatory compliance status
- •Logs lineage metadata only — does not execute or verify the training process
Machine-readable manifest
All package facts are available as a JSON manifest for AI agent consumption.
View developer-products.jsonNeed enterprise support?
Get in touch about enterprise deployments, integrations, and the full HAIEC platform.