AI-powered development workspace with 23 specialized agents for OpenCode. Greek mythology-named agents for planning, TDD, research, code review, and more.
Agent
Adversarial code review orchestrator. Dispatches hunter agents in parallel to find bugs, then requires each finding to be proved by a failing test before reporting. Routes to hunters before "Landing the Plane" as a quality gate. Returns either "Clear to land" or a list of verified failing tests for Vulkanus to fix.
Agent
Analyzes codebase implementation details. Call @codebase-analyzer when you need to find detailed information about specific components. As always, the more detailed your request prompt, the better!
Agent
Locates files, directories, and components relevant to a feature or task. Call @codebase-locator with human language prompt describing what you're looking for. Basically a "Super Grep/Glob/LS tool" - Use it if you find yourself desiring to use one of these tools more than once.
Agent
Finds similar implementations, usage examples, or existing patterns that can be modeled after. Call @codebase-pattern-finder to get concrete code examples based on what you're looking for! It's like @codebase-locator, but it will also give you code details!
Agent
A technical writer who crafts clear, comprehensive documentation. Specializes in README files, API docs, architecture docs, and user guides. MUST BE USED when executing documentation tasks.
Agent
Contextual grep for codebases. Answers "Where is X?", "Which file has Y?", "Find the code that does Z". Fire multiple in parallel for broad searches. Specify thoroughness - "quick" for basic, "medium" for moderate, "very thorough" for comprehensive analysis.
Agent
A designer-turned-developer who crafts stunning UI/UX even without design mockups. Code may be a bit messy, but the visual output is always fire.
Agent
Adversarial hunter for general code review. Reads AGENTS.md to detect convention violations, import pattern violations, logic bugs, naming violations, error handling anti-patterns, and testing practice gaps. Proves AGENTS.md violations and logic bugs with failing .argus.test.ts tests. Uses Static Warning for…
Agent
Adversarial hunter for comment accuracy and quality. Read-only and advisory only. Cross-references every comment claim against actual code implementation. Identifies misleading descriptions, stale TODOs, temporal language, and "what" comments that add no value over the code itself. Emits Static Warnings exclusively …
Agent
Adversarial hunter for security vulnerabilities. Finds auth bypasses, tenant isolation leaks in multi-tenant applications, IDOR vulnerabilities, privilege escalation, missing input validation, and injection risks. Proves each finding with a failing .argus.test.ts test. Uses Static Warning for issues that cannot be…
Agent
Adversarial hunter for silent failure patterns. Finds swallowed errors, empty catch blocks, catch-and-continue, optional chaining hiding failures, and missing error propagation. Proves each finding by writing a failing .argus.test.ts test. Invoked by Argus before "Landing the Plane".
Agent
Adversarial hunter for code simplification. Reads the diff, identifies complexity that can be reduced (nesting, redundancy, unclear names, dead code), applies refactors to production code, then runs ALL existing tests to prove equivalence. If tests pass, the simplification is valid. If tests fail, revert immediately.…
Agent
Adversarial hunter for test coverage gaps. Analyzes the diff to find critical untested paths — error handling, edge cases, negative tests, business logic branches, async failure scenarios. Writes .argus.test.ts files for the missing coverage. INVERTED PATTERN — these tests should PASS (they test correct existing…
Agent
Adversarial hunter for type design and domain invariant violations. Finds anemic types, exposed mutable internals, construction without validation, and missing schema enforcement at boundaries. Proves each finding by writing a failing .argus.test.ts test that demonstrates the invalid state can be created. Invoked by…
Agent
Specialized codebase understanding agent for multi-repository analysis, searching remote codebases, retrieving official documentation, and finding implementation examples using GitHub CLI and Web Search. MUST BE USED when users ask to look up code in remote repositories, explain library internals, or find usage…
Agent
System cartographer and knowledge synthesizer. Maps existing codebase slices by coordinating parallel research probes. Use for "where is X?", "explain the Y system", "what do we know about Z?". Documents what exists without suggesting changes.
Agent
Read-only consultation agent. High-IQ reasoning specialist for debugging hard problems and high-difficulty architecture design.
Agent
Strategic planning consultant. Creates implementation plans through interview-based requirements gathering. Use Tab to switch to Prometheus for complex planning tasks.
Agent
The research equivalent of @codebase-analyzer. Use @thoughts-analyzer when wanting to deep dive on a research topic. Not commonly needed otherwise.
Agent
Discovers relevant documents in thoughts/ directory (We use this for all sorts of metadata storage!). Use @thoughts-locator when you're in a researching mood and need to figure out if we have random thoughts written down that are relevant to your current research task. It's the thoughts equivalent of @codebase-locator.
Agent
Specialized translation agent for accurate, context-aware translations. Handles all translation tasks with strict preservation of formatting, code, and technical terms. MUST BE USED for any translation needs.
Agent
TDD-focused implementation agent. Enforces strict RED → GREEN → VALIDATE → REFACTOR cycle. Works with or without plan files. Delegates specialized work to subagents.
Agent
Master Orchestrator. Routes work to specialists (Prometheus for planning, Vulkanus for implementation, Mnemosyne for research). Never implements directly - delegates everything.