AgentRules.org
RulesStacksClientsAuditorPrunerSelf-HealRFCsBenchmarkCLICI/CDAuthorsSubmit
This specification is maintained by the AgentRules Working Group. Have enhancements or want to claim maintainership?Submit Specification →
devopscritical PriorityID: ebpf-lsm-binary-drift-guardrail

Kernel-Enforced eBPF LSM Container Execution Drift Neutralization

Enforces deterministic, in-kernel blocking of unauthorized executable invocations inside container workloads using Linux Security Modules (LSM) coupled with eBPF hooks. Ensures that process execution drift deviating from cryptographic digest profiles or immutable rootfs manifests is halted prior to execve commitment at ring 0, mitigating zero-day remote code execution vulnerabilities.

Prohibited Anti-Pattern: Relying on User-Space Process Scanners or Post-Execution Audit Logs to Detect Container Compromise

Export Ready-to-Drop Configuration

Select format below to copy code or download file directly.

---
description: Kernel-Enforced eBPF LSM Container Execution Drift Neutralization - Enforces deterministic, in-kernel blocking of unauthorized executable invocations inside container workloads using Linux Security Modules (LSM) coupled with eBPF hooks. Ensures that process execution drift deviating from cryptographic digest profiles or immutable rootfs manifests is halted prior to execve commitment at ring 0, mitigating zero-day remote code execution vulnerabilities.
globs: security/**/*.c, security/**/*.bpf.c, deploy/**/*.yaml, k8s/**/*.yaml, src/security/**/*.rs, src/security/**/*.go
alwaysApply: true
verified_by: "agentrules.org"
license: "CC-BY-4.0"
source: "https://agentrules.org/rules/ebpf-lsm-binary-drift-guardrail"
---

# Kernel-Enforced eBPF LSM Container Execution Drift Neutralization

> **Severity**: CRITICAL | **Category**: devops
> **Anti-Pattern Warning**: Relying on User-Space Process Scanners or Post-Execution Audit Logs to Detect Container Compromise

## System Constraints
You are a Principal Security Architect specializing in Linux kernel instrumentation and cloud-native runtime security. Your mandate is to enforce absolute execution boundaries within container environments using eBPF LSM (Linux Security Modules) hooks. Every containerized binary execution must be validated at ring 0 prior to task credential transition, rejecting any process invocation not explicitly registered in an authorized immutable task manifest.

## Strict Guidelines
- Engineers MUST ALWAYS attach eBPF execution control programs to the `bprm_creds_for_exec` or `bprm_check_security` LSM hooks rather than `sys_enter_execve` tracepoints to prevent race conditions (TOCTOU) and enforce kernel-level execution denial.
- Engineers MUST NOT depend on user-space telemetry agents, ptrace hooks, or synchronous seccomp notifications to prevent zero-day RCE payloads, as user-space evaluation introduces fatal latency windows and evasion vectors.
- Code MUST resolve container cgroup namespaces via `bpf_get_current_cgroup_id()` or equivalent BPF core helpers and cross-reference inode security attributes with pre-populated, cryptographically signed kernel BPF maps before returning 0 (EPERM denial).
- Services MUST emit asynchronous ring-buffer audit events (`BPF_MAP_TYPE_RINGBUF`) with process credential snapshots (UID, GID, PID, comm, executable path) when an execution attempt is blocked, without blocking the kernel execution path.
- Components NEVER allow interactive shells (`sh`, `bash`, `dash`), utilities (`curl`, `wget`, `nc`, `python`), or dynamic shared-library execution within immutable production workloads unless declared in an explicit allowlist manifest.

## Bad vs Good Implementations
### Kernel-Space eBPF LSM Execution Blocker

#### Bad Antipattern (AI Common Hallucination)
> **Why it fails**: Kprobes and syscall tracepoints run as passive observers. Relying on them for runtime enforcement allows unauthorized binaries to fully execute before user-space or downstream filters can respond, completely neutralizing defense against in-memory zero-day exploits.
```c
// Flawed: Monitoring sys_enter_execve using kprobe/tracepoint.
// Tracepoints are passive audit probes; they cannot return -EPERM to halt execution.
SEC("tracepoint/syscalls/sys_enter_execve")
int trace_execve(struct trace_event_raw_sys_enter *ctx) {
    char comm[16];
    bpf_get_current_comm(&comm, sizeof(comm));
    // Cannot deny execution here: the syscall will continue!
    bpf_printk("Executing process: %s\n", comm);
    return 0;
}
```

#### Production Standard (Correct)
> **Standard**: The eBPF LSM hook `bprm_check_security` enforces an atomic kernel-space decision. Returning `-EPERM` instantly terminates binary loading prior to memory page mapping, neutralizing unauthorized binary drift with zero user-space latency.
```c
// Production: LSM hook directly halts execution via -EPERM.
#include <vmlinux.h>
#include <bpf/bpf_helpers.h>
#include <bpf/bpf_tracing.h>

#define EPERM 1

struct {
    __uint(type, BPF_MAP_TYPE_HASH);
    __uint(max_entries, 10240);
    __type(key, u64); // cgroup_id + binary path hash
    __type(value, u8); // 1 = allowed
} allowed_binaries SEC(".maps");

SEC("lsm/bprm_check_security")
int BPF_PROG(restrict_exec, struct linux_binprm *bprm) {
    u64 cgroup_id = bpf_get_current_cgroup_id();
    
    // Read file path hash or check cgroup scope
    u64 lookup_key = cgroup_id;
    u8 *is_allowed = bpf_map_lookup_elem(&allowed_binaries, &lookup_key);
    
    if (!is_allowed) {
        // Deny execution immediately inside the kernel
        return -EPERM;
    }
    
    return 0; // Authorized
}

char LICENSE[] SEC("license") = "GPL";
```
Recommended MCP Servers (mcp-serverlist.com)

Equip agent with system tools to verify code for this stack.

Kubernetes Cluster Controller MCP
Enables inspection and mutation of Kubernetes DaemonSets, SecurityContexts, and runtime policies.
Recommended Agent Skills (askills.dev)

Ready-to-use executable skill functions tailored to this stack.

BPFTool Inspection and Loading Utility
Validates eBPF bytecode loading, inspects active LSM probes, and tests kernel map state in test clusters.
Architectural Specification & Invariant Rationale

Architectural Invariant Design: Kernel-Enforced eBPF LSM Container Execution Drift Neutralization

Kernel-Enforced eBPF LSM Container Execution Drift Neutralization defines an industrial-grade behavioral standard for devops engineering. Enforces deterministic, in-kernel blocking of unauthorized executable invocations inside container workloads using Linux Security Modules (LSM) coupled with eBPF hooks. Ensures that process execution drift deviating from cryptographic digest profiles or immutable rootfs manifests is halted prior to execve commitment at ring 0, mitigating zero-day remote code execution vulnerabilities. Modern software architectures require strict adherence to deterministic patterns to guarantee high availability, maintainability, and zero security vulnerabilities.

Framework Paradigm Shift & Drift MechanicsLLMs are trained on millions of open-source repositories containing outdated code, legacy idioms, and unpatched antipatterns. When coding assistants generate implementations for this stack, they default to common statistical tokens: Relying on User-Space Process Scanners or Post-Execution Audit Logs to Detect Container Compromise Without an explicit system instruction file, developers spend hours debugging silent failures.
Failure Modes & Production Blast Radius

Hazard Analysis: What Breaks Without This Rule

The following failure modes have been cataloged from hundreds of hours of production code reviews where AI assistants produced non-compliant implementations:

Failure ModeAI Symptom (The Anti-Pattern)SeverityProduction Impact & Remedy
AI Hallucination & Deprecation DriftRelying on User-Space Process Scanners or Post-Execution Audit Logs to Detect Container CompromiseCritical
Violating this industrial standard introduces regression bugs, security risks, and unexpected runtime failures in production.
Enforce explicit behavioral system directives across all AI coding assistants in your repository.
Silent Architecture InconsistencyAI assistants produce mixed paradigms across files, creating fragmented codebase conventions.Medium
Increases cognitive load during code reviews and leads to maintenance debt across engineering teams.
Pin project conventions in .cursor/rules, CLAUDE.md, and AGENTS.md so all team agents follow identical guardrails.
Architectural Directives in Detail

Step-by-Step Production Directives

Each rule constraint below is injected into the AI agent's system prompt to establish deterministic boundaries:

1

Engineers MUST ALWAYS attach eBPF execution control programs to the `bprm_creds_for_exec` or `bprm_check_security` LSM hooks rather than `sys_enter_execve` tracepoints to prevent race conditions (TOCTOU) and enforce kernel-level execution denial.

Engineers MUST ALWAYS attach eBPF execution control programs to the `bprm_creds_for_exec` or `bprm_check_security` LSM hooks rather than `sys_enter_execve` tracepoints to prevent race conditions (TOCTOU) and enforce kernel-level execution denial.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
2

Engineers MUST NOT depend on user-space telemetry agents, ptrace hooks, or synchronous seccomp notifications to prevent zero-day RCE payloads, as user-space evaluation introduces fatal latency windows and evasion vectors.

Engineers MUST NOT depend on user-space telemetry agents, ptrace hooks, or synchronous seccomp notifications to prevent zero-day RCE payloads, as user-space evaluation introduces fatal latency windows and evasion vectors.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
3

Code MUST resolve container cgroup namespaces via `bpf_get_current_cgroup_id()` or equivalent BPF core helpers and cross-reference inode security attributes with pre-populated, cryptographically signed kernel BPF maps before returning 0 (EPERM denial).

Code MUST resolve container cgroup namespaces via `bpf_get_current_cgroup_id()` or equivalent BPF core helpers and cross-reference inode security attributes with pre-populated, cryptographically signed kernel BPF maps before returning 0 (EPERM denial).

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
4

Services MUST emit asynchronous ring-buffer audit events (`BPF_MAP_TYPE_RINGBUF`) with process credential snapshots (UID, GID, PID, comm, executable path) when an execution attempt is blocked, without blocking the kernel execution path.

Services MUST emit asynchronous ring-buffer audit events (`BPF_MAP_TYPE_RINGBUF`) with process credential snapshots (UID, GID, PID, comm, executable path) when an execution attempt is blocked, without blocking the kernel execution path.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
5

Components NEVER allow interactive shells (`sh`, `bash`, `dash`), utilities (`curl`, `wget`, `nc`, `python`), or dynamic shared-library execution within immutable production workloads unless declared in an explicit allowlist manifest.

Components NEVER allow interactive shells (`sh`, `bash`, `dash`), utilities (`curl`, `wget`, `nc`, `python`), or dynamic shared-library execution within immutable production workloads unless declared in an explicit allowlist manifest.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
Technical Questions & Answers

Frequently Asked Technical Questions

Common questions developers face when configuring AI coding assistants for this stack:

Q:Why is Kernel-Enforced eBPF LSM Container Execution Drift Neutralization critical for AI coding assistants?

Enforces deterministic, in-kernel blocking of unauthorized executable invocations inside container workloads using Linux Security Modules (LSM) coupled with eBPF hooks. Ensures that process execution drift deviating from cryptographic digest profiles or immutable rootfs manifests is halted prior to execve commitment at ring 0, mitigating zero-day remote code execution vulnerabilities. Without explicit behavioral boundaries, AI coding assistants (Cursor, Claude Code, Windsurf, Codex) rely on training data averages, which frequently produce deprecated, inefficient, or vulnerable code patterns.

Q:How does adopting this rule prevent AI hallucinations?

This rule explicitly declares what the AI must avoid (Relying on User-Space Process Scanners or Post-Execution Audit Logs to Detect Container Compromise) and prescribes verified positive examples. Grounding the LLM with negative and positive pairs eliminates guesswork.

Q:Which AI coding environments are supported?

This specification is multi-platform compatible: Cursor (.cursor/rules/*.mdc & .cursorrules), Anthropic Claude Code (CLAUDE.md), Windsurf (.windsurfrules), Codex / Cline (AGENTS.md), and GitHub Copilot (.github/copilot-instructions.md).

Q:Can I combine this rule with other technology stacks?

Yes. Use the Interactive Stack Builder or download an Agent Pack to merge this rule with other frontend, backend, database, security, and DevOps rules without syntax collision.

Open Specification · CC BY 4.0·Free for Personal, Team & Commercial In-Repo Use

Citation & Open Source Attribution Guidelines

This specification is licensed under Creative Commons Attribution 4.0 International (CC BY 4.0). You are completely free to adapt and share this rule in your workspace. If citing, including in public AI benchmark datasets, or redistributing, attribution to AgentRules.org is required.

Related DEVOPS Rules

View all rules →
devops
86

Kubernetes Gateway API v1 HTTPRoute Canary and ReferenceGrant Invariants

Enforces strict conformance to Kubernetes Gateway API v1 specifications during migration from legacy Ingress controllers. Mandates explicit ReferenceGrant resources for cross-namespace routing, strict relative integer weighting for HTTPRoute canary releases, and explicit backendRef port declarations.

devops
118

ArgoCD ApplicationSet Matrix Generator Invariants & Cascade Deletion Guards

Enforces strict structural constraints on ArgoCD ApplicationSet Matrix generators combining cluster selectors and Git directories. Prevents catastrophic multi-cluster cascade pruning, accidental secret exposure through parameter interpolation, and runaway Application CR explosion via explicit sync policies, preservation finalizers, and parameter validation.

devops
40

Karpenter Resilient Topology, Interruption Queue, and Consolidation Budget Standard

Enforces deterministic disruption budgets, multi-instance-type diversification, and event-driven SQS interruption queue handling across Karpenter NodePools and EC2NodeClasses. Eliminates cascading eviction loops, instance starvation, and availability downtime during AWS Spot rebalances and underutilized node consolidation.

1 Rule in Stack

Kernel-Enforced