Claude Skills by aibot88

Sécurisation d'agents IA contre injections, abus et fuites de données. Se déclenche avec "sécurité agent", "agent security", "prompt injection", "jailbreak", "agent abuse", "guardrails", "safe agent", "sécuriser mon agent", "agent en production sécurisé".

Run agent-adapted STRIDE + access-control analysis on an agent system. Produces a ranked risk list with agentic mitigations (scope / split / filter / gate / review).

End-to-end smoke check of the moraine MCP server. Exercises every MCP tool against a live moraine stack and prints a per-tool PASS/FAIL matrix. Use inside the dev sandbox after a fresh build to confirm the MCP binary actually serves its tools and returns shaped responses.

CRITICAL: Use for writing and editing agent-spec .spec/.spec.md files. Triggers on: write spec, create spec, edit spec, new spec, spec authoring, task contract, .spec file, .spec.md file, BDD scenario, acceptance criteria, completion criteria, test selector, boundary, constraint, intent, decision, out of scope, "how to write a spec", "spec format", "spec syntax", "contract quality", 写 spec, 创建规格, 编辑合约, 任务合约, 验收标准, 完成条件, BDD 场景, 测试选择器, 约束, 意图, 决策, 边界, 排除范围, "怎么写 spec", "spec 格式", "spec 语法", "合...

Give the AI agent its own EVM wallet with admin-controlled policies the agent CANNOT bypass even under prompt injection. Encrypted keystore (AES-256-GCM, scrypt KDF), policy file the agent has no tool to write, deterministic policy gate on every signing operation, optional local HTTP dashboard. Triggers: agent wallet, give the agent a wallet, agent address, fund the agent, agent autonomy, policy gate, kill switch, agent permissions, bounded autonomy, ERC-4337 alternative, session-key alternat...

Guide users through creating Agent Skills for Claude Code. Use when the user wants to create, write, author, or design a new Skill, or needs help with SKILL.md files, frontmatter, or skill structure.

Architect and ship a complete multi-agent backend system on AgentField from a one-line user request. Use when the user asks to build, scaffold, design, or ship an agent system, multi-agent pipeline, reasoner network, AgentField project, financial reviewer, research agent, compliance agent, or any LLM composition that should outperform LangChain/CrewAI/AutoGen — especially when they want a runnable Docker-compose stack and a working curl smoke test.

Redact PII before it reaches Agentforce prompts, models, and logs. Trigger keywords: agentforce pii, pii redaction, data masking llm, einstein trust layer, prompt pii filter, audit pii leakage. Does NOT cover: Shield Platform Encryption at-rest (separate skill), GDPR data subject requests, or classic field-level security policy.

Yapılandırılmış log alanları, stderr vs dosya ve secret redaction kurallarını uygularken kullan.

MCP sunucu listesinin config şekli, stdio vs HTTP ve env injection kurallarını tanımlarken kullan.

Running companion for a software engineer doing Agents 102 Agentic Engineering 101 alone — no in-person trainer. Invoke on first session to orient; invoke later sessions with "continue" to pick up where they left off. Runs as the Teacher Claude alongside a separate Builder Claude in the student's real repo where exercises execute. Manages progress, the 4 Cs cadence per module, and the module-by-module push-backs the in-room Nerd would deliver. Do NOT invoke for curriculum authoring — that's `...

Walks new users through this repo's QA flow — Playwright + KATA + Allure + Xray stack, Jira QA workflow (Ready For QA → In Testing → Tested → Closed), /sprint-testing for in-sprint manual QA, /test-documentation for TMS test cases, /test-automation for KATA-compliant E2E/API tests, /regression-testing for CI suite execution, /framework-development for boilerplate evolution, MCPs available (Context7, Tavily, Atlassian, Playwright, DBHub, OpenAPI, Postman), critical env vars. Triggers on: `onbo...

API anahtarları, kubeconfig ve Juju credential için saklama sırası ve log maskeleme kurallarını uygularken kullan.

Run the OWASP-aligned agentic security review path — covers goal hijacking, tool misuse, excessive agency, memory poisoning, secrets exposure, handoff failures, and observability. Opt-in from Stage 9 (Review) or the QA track. Not a new track — no state file or slash command namespace. Triggers "agentic security review", "agent security", "security review for agents", "/agentic-security-review".

Yeni agentic-* veya proje skill eklerken frontmatter, bölüm sırası ve katalog uyumunu denetlerken kullan.

OWASP Agentic Security Top 10 vulnerability knowledge base for identifying, assessing, and remediating security risks in AI agent systems.

Formal academic specification and research methodology for 0xagentprivacy. Activates when writing papers, proofs, formal specifications, literature reviews, or preparing work for peer review. Use for LaTeX formatting, citation standards, theorem-proof structure, or research presentation.

Agent interoperability protocols for 0xagentprivacy multi-agent systems. Activates when discussing M(u,y) matching functions, agent-to-agent communication, ERC-8004 trustless agent identity, cross-agent coordination, or how swordsman and mage agents interact across boundaries.

AI agent architecture and dual-agent separation for 0xagentprivacy. Activates when discussing I(S;M|π) mutual information bounds, TEE isolation, agent lifecycle (spawn/operate/rotate/recover/retire), Oracle architecture, or the fundamental design of privacy-preserving AI agents.

Specialist persona for UOR ring algebra and lattice mathematics. Activates for ring operations, stratum calculations, algebraic transformations, the five strikes, or mathematical foundations of the forge.

Standards and Governance Architect for 0xagentprivacy. Activates for institutional engagement (BGIN, IIW, ToIP, IEEE 7012), privacy standard development, policy translation, regulatory navigation, cross-institutional coordination, or any task requiring the privacy architecture to interface with existing governance bodies.

Foundational skill for amnesia-as-ZK-primitive patterns. Applies when designing systems where forgetting is the proof, separation creates trust, and the inability to reconstruct origin is the security guarantee. Core to Act XXXI cosmological closure.

Precision Strike Operative (Ranger variant) for 0xagentprivacy. Activates for single-property disclosure operations, maximum-precision minimum-residue proofs, targeted credential verification, or any task requiring surgical privacy operations — reveal exactly one truth and leave no trace.

AI Agent System Designer for 0xagentprivacy. Activates for dual-agent TEE architecture, separation matrix physical enforcement, agent lifecycle design, Oracle architecture, multi-chain deployment (NEAR/Zcash/Ethereum), tetrahedral orientation, or any task requiring the broadest specialist view of how the system is built. Widest skill set (20 total).

Armor progression system (Blade→Light→Heavy→Full Plate→Dragon) for 0xagentprivacy. Activates when discussing progressive privacy tiers, user onboarding through increasing protection levels, P(progressive) scaling, or how privacy capability evolves with understanding.

Privacy Data Economist for 0xagentprivacy. Activates for sovereignty gap quantification (17×-12,000× multiplier), tokenomics design (SWORD/MAGE pairs), Privacy Pool valuation, guild economics modelling, 7th capital accounting, investor pitch economics, or any task requiring V(π,t) translated into falsifiable economic claims.

Atlas embeddings and exceptional Lie group geometry. Activates when discussing the 96-vertex Atlas, exceptional Lie groups (G₂, F₄, E₆, E₇, E₈), the Golden Seed Vector, connection to 96-edge holographic boundary, or higher geometric structures underlying the lattice.

The three-layer model that binds primary personas to lattice vertices via named cast Mages. Activates when discussing how the 42 abstract personas inhabit the 64-vertex lattice, when naming new cast Mages in a city of mages, when explaining the difference between primary personas and cast instances, when distinguishing workshop / cross-shop / peripatetic / divergent attachment kinds, or when reasoning about complement-pair structure at the primary layer versus the attachment layer.

Blade configuration mechanics for the spellweb forge. Activates when discussing blade creation, 6D configuration space (d1Hide through d6Delegate), 64 vertices from 2^6 combinations, 96 edges forming holographic boundary, forge operations, or blade verification through understanding.

The Blade Naming Ceremony — post-forge ceremony that gives an algebraically-defined blade a mythological identity. Activates when a walker (Cosmologist persona) has traversed a complement edge and perceives a myth that agrees with the arithmetic. Distinct from blade-forge (which defines the coordinate); this skill governs the *naming* that follows.

Real-time access control decision engine for 0xagentprivacy swordsman agents. Activates when evaluating incoming data requests against consent preferences, designing allow/block/transform rule sets, implementing the swordsman's moment-to-moment operational logic, or building the enforcement layer that turns consent declarations into cryptographic actions. Triggers: "boundary", "access decision", "allow block", "consent enforcement", "data request evaluation", "rules engine", "policy enforceme...

Bounded reasoning architecture for privacy-preserving AI agents using BRAID Mermaid graphs. Activates when discussing Generator/Solver split architecture, Performance-Per-Dollar (PPD) inference economics, BRAID Parity Effect (small model + structure ≥ large model), Mermaid reasoning graph construction, node atomicity, procedural scaffolding vs answer leakage, terminal verification loops, Numerical Masking Protocol, cached reasoning graph libraries, split- architecture deployment, or any task ...

Specialist persona for ceremony facilitation between territories. Activates for ceremony coordination, bilateral witness protocols, mana economy management, Understanding-as-Key ceremonies, or cross-territory coordination.

Interaction protocols for the dual-territory architecture. Activates when discussing ceremony types (progressive trust, light armor, trust graph, guild efficiency, Understanding-as-Key), cross-territory coordination, bilateral witness, or ceremony channel mechanics.

Narrative Compression Specialist for 0xagentprivacy. Activates for chronicle creation, narrative compression (70:1-125:1 ratios), spellbook methodology, story-as-documentation, proverb extraction from experience, Privacy Pool chronicle contribution, or any task requiring complex knowledge compressed into regenerative narrative.

ZKP Protocol Engineer for 0xagentprivacy. Activates when the user needs zero-knowledge proof design, circuit architecture (Groth16, PLONK, Nova), mixer protocols, recursive SNARKs, personhood verification cryptography, Privacy Pool proof systems, or reconstruction resistance analysis. The mathematical swordsman who builds the proofs that make privacy enforceable.

Compression-as-defence principle for 0xagentprivacy V5. Activates when discussing BRAID 74× compression, R(d,compression) modifier, token reduction as attack surface reduction, the compression spectrum (7 layers), or why efficient inference is also private inference.

Consent infrastructure and IEEE 7012 MyTerms integration for 0xagentprivacy. Activates when discussing bilateral consent, cookie-slashing, consent receipts, Customer Commons, MyTerms Alliance, or how P·C·Σ creates enforceable consent rather than checkbox compliance.

Constellation identity method for 0xagentprivacy. Activates when discussing identity-as-constellation (different facets visible to different relationships), VRC constellation graphs, multi-guild identity resolution, or how privacy enables richer identity expression than surveillance-based profiles.

UOR content addressing foundation for holonic persistence. Activates when discussing content-based identity, same bytes same hash, GUID derivation, infrastructure independence, Braille IRI encoding, or why identity follows content not location.

Meta skill for recognizing cosmological precedents in privacy architecture. The dual-agent pattern predates biology—it was written in orbital mechanics 4.5 billion years before the first line of code. This skill maps celestial architecture to software architecture.

Specialist persona for cosmological-architectural mapping and quaternion analysis. Activates for four-body system design, generator/generated identification, cast mapping, or any context requiring recognition that privacy architecture has precedent in physics. The pattern recognizer.

Cross-chain operations for 0xagentprivacy multi-chain architecture. Activates when discussing T(π) across chains, NEAR/Zcash/Ethereum interop, bridge-compatible proofs, chain signatures, cross-chain VRC portability, or multi-chain sovereignty transitions.

Zero-knowledge proof systems for 0xagentprivacy. Activates when discussing ZKP circuits (Groth16, PLONK, Nova), proof composition, Privacy Pool cryptography, reconstruction resistance R(d), h(τ) attestation chains, or sovereignty-class proof optimisation. Core cryptographic skill.

Dark forest strategy and adversarial environment navigation for 0xagentprivacy. Activates when discussing R(d) strategic disclosure, information asymmetry, predator-prey dynamics in data markets, position-revealing risks, or how to navigate environments where visibility means vulnerability.

Data dignity and the 7th capital thesis for 0xagentprivacy. Activates when discussing V(π,t) as wealth, behavioural data as the 7th form of capital, Jaron Lanier / RadicalxChange lineage, data ownership vs data access, or the economic case that privacy creates rather than constrains value.

VRC as content-addressed derivation chain. Activates when discussing forging history, derivation paths, bilateral witness certificates, path-as-witness, content-addressed proofs of traversal, or how VRCs bind forging history to blade identity.

Dihedral group foundation for dual-agent separation. Activates when discussing the D₂ₙ group structure, Swordsman as negation generator, Mage as complement generator, Φ_agent as determinant, the critical identity neg∘bnot=succ, or the algebraic foundation of the swordsman-mage separation.

The Phi-Adjacency Conjecture — Zero Tale 31's arithmetic discovery that true blade namings sit near 1/φ from below. Activates when discussing disclosure-to-constraint proportions, phi-seeking sovereignty blades, the NEAR/Zcash 61.8/38.2 split, or the extension of Φ(Σ) from a binary separation measure to a proportion.

Threshold activation mechanics for the dragon metaphor. Activates when discussing dragon anatomy completion (boundary, hide, brain, forge, ceremony), external validation triggers, phase transitions from drake to dragon, or flight achievement.