Skip to main content

Tool Ecosystem

In one line: Select tools by model capability and development phase, not by habit — this section is the decision matrix.

The methodology integrates plugins, MCP servers, and skills into a unified stack. Tool selection is capability-based.

Capability Assessment

Before selecting tools, assess the AI model's capabilities:

CapabilityHow to AssessThreshold
Context windowCheck model documentationLarge (>200K), Medium (50-200K), Small (<50K)
Internal reasoningDoes the model have thinking/reasoning blocks?Yes / No
Tool reliabilityCan the model use tools (Read, Edit, Bash) without errors?Reliable / Unreliable
Multi-step planningCan the model decompose and execute 5+ step plans?Yes / Needs guidance
Code understandingCan the model read a 500-line file and modify precisely?Full context / Needs navigation

Tool Recommendations by Capability

Code Navigation

CapabilityRecommendation
Reliable direct inspection within the task budgetRead scoped source and configuration; preserve authoritative references.
Cross-file navigation or dependency analysisEvaluate approved symbol/graph tools such as Serena or codebase-memory-mcp; verify index freshness and coverage.
Context pressure or incomplete retrievalReduce/split scope or use a tested retrieval binding; do not infer that a named plugin is mandatory from context-window size.

Reasoning Support

CapabilityRecommendation
Planning and difficult decisionsEvaluate observable task outcomes, counterexamples and concise decision rationale; visible reasoning blocks do not determine correctness.
Optional decomposition toolsAdopt only where they improve measured work within approved data/tool boundaries; they are not an audit prerequisite.
Audit trail requiredStore concise rationales, evidence references, relevant outputs and approval events under the applicable retention policy; not hidden reasoning chains.

Code Quality

LayerToolWhen
Real-timeTypeScript LSP, Pyright LSPDuring implementation
Pre-commitruff (lint), pyright (types), tsc (TypeScript)Every commit
SecurityAikido plugin (SAST, secrets)During development
PR reviewApproved code-review integrationBefore merging, on the actual candidate
PR securityclaude-code-security-review GitHub ActionOn every PR
Post-mergecode-simplifier (bloat detection)After major features

These are reference-host examples, not universal vendor or cadence requirements. The approved stack/evidence profile selects actual tools, versions, subject scope, source-processing permissions and required execution. A package name or model label does not establish installation or completion.


Tool Recommendations by Development Phase

Brainstorm Phase

ToolPurposeCapability Requirement
Visual companion (browser)Mockups, diagramsAny
Terminal discussionConceptual choicesAny
Architecture index lookupUnderstand existing codeAny

Existing-system brainstorming often needs source and integration discovery. Use authorized read-only evidence where it answers the question; a new-product discussion may need none.

Design Phase

ToolPurposeCapability Requirement
Architecture indexFind relevant doc pagesAny
Serena get_symbols_overviewUnderstand existing interfacesMedium/small context
Read toolRead existing files directlyLarge context

Plan Phase

ToolPurposeCapability Requirement
Architecture index tests: fieldIdentify test filesAny
Architecture index component_mapMap files to tasksAny

Implement Phase

ToolPurposeCapability Requirement
Read / Edit / BashDirect file manipulationLarge context, reliable tools
Serena find_symbol / replace_symbol_bodyPrecise symbol editingMedium/small context
TypeScript LSPFrontend type checkingAny (if available)
Context7Library documentation lookupAny (prevents hallucinated APIs)
Temporal MCPWorkflow debuggingProjects using Temporal
Neo4j MCPGraph queriesProjects using Neo4j

Review Phase

ToolPurposeCapability Requirement
code-review pluginScoped candidate reviewApproved model/tool binding
Domain-specific review agentsCompliance, security, API, migrationProject-specific
CodeRabbitExternal AI review perspectiveOptional; approved source processing and current service terms

Verify Phase

ToolPurposeCapability Requirement
Stop hookDebug-only checklist in the reference adapterVerify host delivery separately
Architecture index tests:Declared mapping; compare against actual executionA mapping alone is not a receipt
ruff / pyright / tscLint + type checkAny
AikidoSecurity scanAny (if configured)

All phase tables are illustrative capability mappings. The general architecture index is an adopter extension, not a supplied plugin; tests metadata is not an execution receipt. Context thresholds, pricing, package names and model selections must be verified for the actual binding rather than treated as fixed acceptance rules.


Plugin Stack Reference

Reference lifecycle and review tools

PluginAdoption referencePurpose
SuperpowersVerify package, version and host procedure under §4.4Lifecycle skills
code-reviewSelect the approved review integrationScoped agent-assisted review
code-simplifierOptional, evaluated helperRefactoring suggestions requiring verification
PluginAdoption referenceWhen
typescript-lspVerify approved host packageProjects selecting TypeScript language services
pyright-lspVerify approved host packageProjects selecting Python language services
SerenaVerify approved host package and indexTasks benefiting from symbol navigation

MCP Servers (configure based on stack)

ServerPackageWhen
Context7Verify approved provider/packageOptional library-documentation retrieval
Temporal MCPtemporal-mcpProjects using Temporal
Neo4j MCP@johnymontana/neo4j-mcpProjects using Neo4j
codebase-memory-mcpcodebase-memory-mcpLarge codebases, token optimization
Sequential Thinkingsequential-thinkingOptional structured problem decomposition; not a prerequisite for an audit trail

Cross-Tool Compatibility

AGENTS.md

Use the selected host's supported instruction entry point and verify actual loading. Where two hosts support the same content, an approved pointer or symlink may avoid duplication; do not overwrite an existing AGENTS.md or assume every host follows links and precedence identically. Preserve tool-specific differences as explicit, bounded references.

Adoption Tier Impact

TierCapability obligationBinding choice
CoreAdopted authority, lifecycle, evidence and required controlsExisting approved tools or verified reference adapters; no universal vendor install list
RecommendedNavigation, language services and domain integration where usefulEvaluate against workload, permissions, privacy and maintenance cost
ExtendedAdditional automation/retrieval where justifiedExpand only with scoped acceptance evidence and approved operating authority

Competitive Landscape (March 2026)

The former comparison table and uniqueness claim did not establish a defined comparison corpus or independent outcomes. They are not a basis for adoption. S4U's intended value is a maintained combination of business-intent traceability, bounded engineering procedure and inspectable evidence. Evaluate that value locally and compare alternatives using the factory research; no market-wide superiority is claimed.