Subagents with their own prompt and tools. Injected whole, so what you see in the token column is what every session pays.
Agent Claude Code
Deep expert on SCIP (Solving Constraint Integer Programs) and MINLP solver architecture — spatial branch-and-bound, constraint handlers, separators/cuts, presolve, expression/nonlinear handling, grounded in the SCIP Optimization Suite source. Use when comparing discopt against SCIP, reasoning about MINLP solver…
felipefontoura/spec-driven-development-book
Agent Claude Code
Produces requirements.md and design.md from an approved idea or feature request. Use for the WHAT and the HOW — user stories, EARS requirements, data models, API contracts, technical decisions with rationale, and the requirement→design traceability map. Never writes implementation code.
Agent Claude Code
Opus adversarial reviewer for a proposed DSL shape. Given the kroot-flesh-shaper proposal, kroot-lone-spear coverage, and kroot-trail-shaper describer spec, it attacks the shape on four axes — sprawl (an existing shape already covers this), flattening (adopting it distorts a family member's meaning), fidelity/parity…
girijashankarj/cursor-handbook
Agent Claude CodeCursor
Writes comprehensive tests and improves test coverage using project testing framework.
10Legs/freelance-developer-harness
Agent Claude Code
Use when auditing the health of a client repo's version control (branches, PRs, commits, tags), when advising on branching strategy or commit conventions, when a developer asks "should I branch for this?", when checking for stale branches or unreviewed PRs before a sprint, when preparing a release or tag, or when you…
Agent Claude Code
This agent operates under library-first constraints.
ocbunknown/fastapi-claude-template
Agent Claude Code
Use proactively after non-trivial changes in src/ to review architecture compliance (layer rules, requestbus/DBGateway patterns from CLAUDE.md) AND security/authorization correctness — verify that admin-only parameters cannot be supplied by regular users, that user-facing responses don't leak internal/business data…
Agent Claude Code
Compresses the project memory hot layer when it grows too large. Reads MEMORY.md and schema.yaml, archives older entries into archive.md, and rewrites MEMORY.md keeping only recent high-value entries. Invoke with /memory-loop:compress or when check-compress warns about file size.
Agent Claude Code ✓ vendor
Evaluates code changes for correctness, style adherence, architecture alignment, testing coverage, and documentation completeness.
Agent Claude Code
Schema markup specialist detecting, validating, and generating structured data (JSON-LD preferred). Focuses on schemas that improve AI discoverability including Organization, Person, Article, sameAs, and speakable properties.
Agent Claude Code
Platform optimization specialist analyzing readiness for Google AI Overviews, ChatGPT web search, Perplexity AI, Google Gemini, and Bing Copilot.
Agent Claude Code
MySQL database specialist for query optimization, schema design, security, and performance. Use PROACTIVELY when writing SQL, creating migrations, designing schemas, or troubleshooting database performance.
PolicyEngine/policyengine-claude
Agent Claude Code
Stage-5 comparator. Takes the microsim output and the prior-scores anchor, returns PASS (within expected band, write up the result) or INVESTIGATE (mismatch, trigger calibration-diagnostics). Normalizes single-year vs 10-year scores and accounts for dataset-version drift.
Agent Claude Code
Petra - The Wordsmith ✍️. Prompt engineering specialist who crafts instructions that work with LLM mechanics. Invoke when creating agent definitions, system prompts, or LLM instructions. Leverages token prediction, attention mechanisms, and pattern reinforcement for maximum effectiveness.
danielrosehill/Claude-Code-Plugins
Agent Claude Code
Refactor an existing Daniel Rosehill Claude Code workspace repo into a shareable Claude Code plugin. Fetches the source repo, extracts reusable slash commands / agents / skills, depersonalises content, and scaffolds a new plugin repo under /repos/github/my-repos/claude-code/cc-plugins/plugins/ /. Does NOT push to…
german-krasnikov/unity-biome-mcp
Agent Claude Code
Use after implementation to verify Unity Play Mode behavior with data assertions, NUnit, or playtest DSL. Do not use for scene authoring, code changes, or screenshot-only acceptance.
Agent Claude Code
A project-management agent for chip-design work that tracks requirements, risks, progress, milestones, interfaces, and quality gates. It also consults a built-in technical knowledge base for design decisions.
Agent Claude Code
Principal automotive SE specializing in ISO 26262 ASIL allocation, AUTOSAR Classic/Adaptive architecture, ISO 21434 cybersecurity, SOTIF, and multi-tool MBSE crosswalks for ADAS and powertrain programs.
nikandr-surkov/ai-saas-starter
Agent Claude Code
Read-only reviewer for auth, billing, and credits-ledger changes. Use proactively after modifying src/lib/auth, src/lib/billing, src/lib/credits, or src/app/api — before committing.
Agent Claude Code
Synthesizes all reviewer reports into a unified editorial decision letter and revision roadmap.
Agent Claude Code
AI DevOps agent for Harness CI/CD platform.
erikdarlingdata/claude-plugins
Agent Claude Code
Fixes one issue, or a small batch of RELATED issues in the same files or subsystem, in its own git worktree, and opens a draft PR. Needs a self-contained brief; does not investigate open-ended scope or replan the queue.
Agent Claude Code needs its repo
Specialist for US-stock industry-position and competitive-moat review — Porter Five Forces scan, market-share trend (absolute and relative), TAM size and trajectory, unit economics, moat classification (7 named types), substitute and new-entrant threats, pricing-power evidence (cross-cycle margin stability)…
Agent Claude CodeCursor
Implements Accepted specs end to end, fixes bugs, and refactors without changing behaviour. Use proactively for new features once the spec is Accepted, domain/Prisma/Server Action changes, and technical approach decisions. Do not use when the request is a problem without a spec — that is product-manager.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: