README

---

Library-First Directive

This agent operates under library-first constraints:

1. Pre-Check Required: Before writing code, search:

   • .claude/library/catalog.json (components)
   • .claude/docs/inventories/LIBRARY-PATTERNS-GUIDE.md (patterns)
   • D:\Projects\* (existing implementations)

2. Decision Matrix:

   Result	Action
   Library >90%	REUSE directly
   Library 70-90%	ADAPT minimally
   Pattern documented	FOLLOW pattern
   In existing project	EXTRACT and adapt
   No match	BUILD new

---

---

---

STANDARD OPERATING PROCEDURE

Purpose

• Mission: Legacy description preserved in appendix.
• Category: README.md; source file: README.md
• Preserve legacy directives (see VCL appendix) while delivering clear, English-only guidance.

Trigger Conditions

• Activate when tasks require README responsibilities or align with the README.md domain.
• Defer or escalate when requests are out of scope, blocked by policy, or need human approval.

Execution Phases

1. Intake: Clarify objectives, constraints, and success criteria; restate scope to the requester.
2. Plan: Outline numbered steps, dependencies, and decision points before acting; map to legacy constraints as needed.
3. Act: Execute the plan using allowed tools and integrations; log key decisions and assumptions.
4. Validate: Check outputs against success criteria and quality gates; reconcile with legacy guardrails.
5. Report: Provide results, risks, follow-ups, and the explicit confidence statement using ceiling syntax.

Guardrails

• User-facing output must be pure English; do not include VCL/VERIX markers outside the appendix.
• Apply least-privilege tooling; avoid leaking secrets or unsafe commands.
• Honor legacy rules, hooks, and budgetary constraints noted in the appendix.
• For uncertain claims, prefer clarification over speculation and cite evidence when observed.

Output Format

• Summary of actions performed or planned.
• Decisions, assumptions, and blockers.
• Next steps or handoff notes with owners and timelines.
• Confidence statement using the required syntax: "Confidence: X.XX (ceiling: TYPE Y.YY)" with the appropriate ceiling (inference/report 0.70; research 0.85; observation/definition 0.95).

Tooling & Integration

• Model: auto
• Allowed tools: None specified
• MCP/Integrations: Not specified; inherit from runtime defaults
• Memory/Logging: Record key events and rationale when supported.

Validation Checklist

• Trigger conditions matched and scope confirmed.
• Plan and execution steps follow the ordered phases.
• Output includes English-only narrative and explicit confidence ceiling.
• Legacy constraints reviewed and applied where relevant.

VCL COMPLIANCE APPENDIX (Internal Reference)

[[HON:teineigo]] [[MOR:root:P-R-M]] [[COM:Prompt+Architect+Pattern]] [[CLS:ge_rule]] [[EVD:-DI<policy>]] [[ASP:nesov.]] [[SPC:path:/agents]] [direct|emphatic] STRUCTURE_RULE := English_SOP_FIRST -> VCL_APPENDIX_LAST. [ground:prompt-architect-SKILL] [conf:0.88] [state:confirmed] [direct|emphatic] CEILING_RULE := {inference:0.70, report:0.70, research:0.85, observation:0.95, definition:0.95}; confidence statements MUST include ceiling syntax. [ground:prompt-architect-SKILL] [conf:0.90] [state:confirmed] [direct|emphatic] L2_LANGUAGE := English_output_only; VCL markers internal. [ground:system-policy] [conf:0.99] [state:confirmed]

Legacy Reference

<details> <summary>Legacy content (verbatim)</summary> <pre># Agents Directory

Kanitsal Cerceve (Evidential Frame Activation)

Kaynak dogrulama modu etkin.

Status: 211 agents organised into ten functional categories Last reorganised: 2025-11-26 Common format: Every agent uses Markdown with YAML front matter (name, type, phase, category, capabilities, tools, MCP servers, hooks, quality gates, artifact contracts).

Recent additions (2025-11-26):

• specialists/finance/quant-analyst.md - Quantitative trading and signal calibration
• specialists/finance/risk-manager.md - Risk quantification and compliance
• specialists/finance/market-data-specialist.md - Real-time market data integration

---

Quick Category Overview

Category	Description	Agent count	Example agents
delivery	Feature and product implementation specialists covering architecture, backend, frontend, and SPARC execution	18	delivery/development/backend/dev-backend-api.md, delivery/architecture/system-design/arch-system-design.md
foundry	Agent creation, templates, registries, and base specialists that support building new automations	19	foundry/core/base-template-generator.md, foundry/templates/skill-boilerplate-generator.md
operations	DevOps, infrastructure, performance, and monitoring agents that keep systems healthy	28	operations/devops/ci-cd/ops-cicd-github.md, operations/infrastructure/terraform/terraform-provisioner.md
orchestration	Goal planners, swarm coordinators, and consensus agents for multi-agent workflows	21	orchestration/consensus/byzantine-coordinator.md, orchestration/swarm/hierarchical-coordinator.md
platforms	Data, ML, neural, Flow Nexus, and platform service agents	44	platforms/ai-ml/automl/automl-optimizer.md, platforms/flow-nexus/multi-model-orchestrator.md
quality	Analysis, audit, testing, and verification agents	18	quality/analysis/code-analyzer.md, quality/testing/test-orchestrator.md
research	Research, reasoning, emerging tech, and discovery agents	11	research/archivist.md, research/emerging/arvr/ar-vr-developer.md
security	Compliance, pentest, container, and cloud security specialists	5	security/compliance/soc-compliance-auditor.md, security/pentest/penetration-testing-agent.md
specialists	Domain specialists for business, industry, finance, and vertical workflows	18	specialists/business/business-analyst.md, specialists/finance/quant-analyst.md, specialists/supply-chain/logistics-optimizer.md
tooling	Documentation, GitHub, and knowledge tooling agents	24	tooling/documentation/api-docs/docs-api-openapi.md, tooling/github/pr-manager.md

Totals include nested directories; for example platforms/ai-ml/* contributes to the platforms count.

---

Directory Layout

agents/
  delivery/                 # Architecture, development, SPARC execution
  foundry/                  # Core, templates, registries, agent builders
  operations/               # DevOps, infrastructure, monitoring, optimization
  orchestration/            # Coordinators, consensus, goal planners, swarm agents
  platforms/                # Flow Nexus, AI/ML, database, neural services
  quality/                  # Analysis, audits, testing specialists
  research/                 # Research cores, reasoning, emerging technology scouts
  security/                 # Compliance, security, penetration testing
  specialists/              # Business, industry, finance, and other domain experts
    business/               # Business analyst, marketing, sales, product management
    finance/                # Quant analyst, risk manager, market data specialist
    supply-chain/           # Logistics, inventory, procurement
  tooling/                  # Documentation, GitHub, productivity tooling
  registry/                 # (moved to foundry/registry)

Each category folder preserves the prior domain structure (for example delivery/development/backend/*).

---

Agent File Structure

Every agent Markdown file starts with YAML metadata followed by usage documentation. Typical sections include:

---
name: agent-name
type: coordinator | coder | analyst | optimizer | researcher | specialist
phase: planning | development | testing | deployment | maintenance
category: delivery
description: Short mission statement
capabilities: [...]
tools_required: [...]
mcp_servers:
  required:           # MCPs that MUST be enabled for this agent to function
    - memory-mcp      # Always required for cross-session memory
  optional:           # MCPs that enhance functionality but aren't required
    - ruv-swarm       # For swarm coordination
    - flow-nexus      # For cloud features
  auto_enable: true   # If true, Claude Code will prompt to enable missing MCPs
hooks:
  pre: |
    # commands...
  post: |
    # commands...
quality_gates: [...]
artifact_contracts: [...]
---

# Agent Title

## When to use
...

---

Finding the Right Agent

1. Browse by category – open the folder matching your goal (for example, orchestration/ for swarm coordinators).
2. Use the registry – foundry/registry/registry.json lists canonical agent metadata for automation.
3. Search by capability – rg "capabilities:.*<keyword>" agents to locate agents with specific skills.
4. Cross-reference skills – many agents reference complementary skills under skills/ in the same functional category.

---

Memory-MCP Integration

All 211 agents have access to the Memory-MCP triple-layer storage system for persistent memory across sessions.

Configuration

Memory-MCP is configured in identity/memory-mcp-config.json and provides:

Capability	Tier	Latency	Purpose
kv_store	1	<1ms	O(1) preference lookups
vector_search	3	<100ms	Semantic search via embeddings
graph_service	4	<50ms	Entity relationship graphs
event_log	5	<10ms	Temporal event logging
query_router	-	-	Polyglot tier selection

Tagging Protocol (Required)

All memory writes MUST include these tags:

WHO: "{agent-type}:{agent-id}"    # e.g., "code-analyzer:601c545c"
WHEN: "{ISO-8601-timestamp}"      # e.g., "2025-12-28T09:47:04.638Z"
PROJECT: "{project-name}"         # e.g., "memory-mcp-triple-system"
WHY: "{purpose}"                  # e.g., "analysis", "implementation", "bugfix"

Memory Namespaces

Namespace	Pattern	Used For
agents	agents/{category}/{type}/{project}/{timestamp}	Agent-specific memories
swarm	swarm/{coordination-type}/{task-id}	Swarm coordination state
expertise	expertise/{domain}	Domain expertise files
findings	findings/{agent-type}/{severity}	Code analysis findings
decisions	decisions/{project}/{decision-id}	Decision records

Access Levels by Role

Role	Operations	Namespaces
admin	All	All
developer	set, get, delete, search, log_event	agents/, swarm/, findings/*
reviewer	get, search, query_by_date	agents/, findings/, decisions/*
coordinator	All	swarm/, agents/
analyst	get, search, query_by_date, find_path	All (read)

Example Usage

# Store a finding
kv_store.set_json('findings:code-analyzer:high:GOD-001', {
    'WHO': 'code-analyzer:601c545c',
    'WHEN': datetime.now().isoformat(),
    'PROJECT': 'my-project',
    'WHY': 'analysis',
    'content': 'Found God Object with 0.26 cohesion',
    'severity': 'high'
})

# Build knowledge graph
graph.add_relationship('finding:GOD-001', 'fixed_by', 'pattern:facade')

# Log event
event_log.log_event(EventType.QUERY_EXECUTED, {
    'agent': 'researcher',
    'query': 'similar issues',
    'results': 5
})

Agent YAML Configuration

Ensure agents declare memory-mcp in their frontmatter:

mcp_servers:
  required:
    - memory-mcp      # Cross-session memory (always required)
  optional:
    - ruv-swarm       # Swarm coordination
  auto_enable: true

rbac:
  api_access:
    - memory-mcp      # Required for memory operations

---

Updating or Adding Agents

1. Place new agents inside the category that matches their primary function.
2. Follow the YAML + documentation structure above; keep hooks and quality gates up to date.
3. Update this README and any relevant taxonomy docs (see docs/agent-taxonomy/) with counts and locations.
4. Regenerate foundry/registry/registry.json if the agent should appear in the programmatic registry.

This organisation mirrors the skills directory, making it easier to align agents, skills, and SOPs by functional area.

---

Promise: <promise>README_VERIX_COMPLIANT</promise> </pre>

</details>