foundry-security-spec
OfficialBuild agentic AI security evals with Foundry.
Software Engineering#specification#agentic ai#sandboxing#evidence gating#security evaluation#multi-agent architecture#CodeGuard rules
AuthorAradotso
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This skill helps you implement Cisco Foundry’s specification for agentic AI security evaluation systems, ensuring multi-agent coordination, evidence-gated findings, and reproducible security outcomes.
Core Features & Use Cases
- Multi-agent Foundry blueprint: Defines an 8-core-role architecture (plus optional extension roles) for orchestrated discovery, validation, and reporting.
- Evidence-gated finding lifecycle: Enforces “no evidence, no finding” principles with validator confirmation and fingerprint-based deduplication.
- Detectors + explorers flywheel: Uses deterministic rule-based detection (CodeGuard-compatible) and exploratory hypothesis testing to uncover new issues and feed back rule improvements.
- Governance and safety constraints: Covers sandboxing, budgets, coverage gates, heartbeat liveness, and fail-safe defaults for safe evaluation workflows.
Quick Start
Tell your AI to generate a Foundry evaluation system by adopting the constitution, seeding spec.md into your spec-kit workflow, and producing an implementation plan with core roles and evidence gates.
Dependency Matrix
Required Modules
None requiredComponents
Standard package💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: foundry-security-spec Download link: https://github.com/Aradotso/security-skills/archive/main.zip#foundry-security-spec Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.