20 AI agents in your git workflow. Secrets, injection flaws and unsafe migrations are blocked at commit; performance, coverage and the rest are reviewed at push. Research → Plan → Implement workflow with GO/NO-GO gates before you write a line. Generates production-ready code, product wikis and codebase maps on demand. Any language. Free.
This repository also configures its own agents. See what manta tells them →
Command Claude Code
Command "audit" from mantacron/manta, covering review depth, step 0 — load suppressions, step 1 — gather context, shared project map — built once, passed to every agent and step 2 — run agents in parallel.
Command Claude Code
Begin by outputting: [ Manta — Blueprint ].
Command Claude Code
Scan the existing codebase and auto-generate PATTERNS.md with detected conventions. Run once, review the output, then commit it so all agents enforce your team's patterns.
Command Claude Code
Begin by outputting: [ Manta — Debt Ledger ].
Command Claude Code
Begin by outputting: [ Manta — Explain ].
Command Claude Code
Command "fix" from mantacron/manta, covering usage, instructions, step 1: choose the report, step 2: extract findings and step 3: run remediation-agent.
Command Claude Code
Begin by outputting: [ Manta — Generate Tests ].
Command Claude Code
Command "init" from mantacron/manta, covering /init — project initialization, mode selection — ask this first, spec-driven init wizard (mode a), before starting: check for saved session and before starting: check existing state.
Command Claude Code
Begin by outputting: [ Manta — POC Init ].
Command Claude Code
Begin by outputting: [ Manta — Pre-Commit Review ].
Command Claude Code
Begin by outputting: [ Manta — Pre-Push Review ].
Command Claude Code
Begin by outputting: [ Manta — Review · 4 agents ].
Command Claude Code
RPI Step 4 — Execute phased implementation with validation gates. Reads PLAN.md and eng.md, implements phase-by-phase, runs test gates after each phase, and writes an IMPLEMENT.md record.
Command Claude Code
RPI Step 3 — Generate comprehensive planning documentation for a feature that received a GO. Produces pm.md (product requirements), ux.md (UX design), eng.md (engineering spec), and PLAN.md (phased implementation roadmap).
Command Claude Code
RPI Step 2 — Research and analyze a feature for GO/NO-GO. Runs 6 agents in sequence (requirement-parser → product-manager → Explore → senior-software-engineer → constitutional-validator → technical-cto-advisor) and writes rpi/{slug}/research/RESEARCH.md.
Command Claude Code
Begin by outputting: [ Manta — Scaffold ].
Command Claude Code
Begin by outputting: [ Manta — Security Scan ].
Command Claude Code
Convert design files (screenshots, Figma exports, wireframes) into production-ready UI components matching the project's conventions. Reads from ui-designs/ by default. Enforces DRY by detecting existing components to reuse. Generates responsive, accessible, typed components with tests and optional Storybook stories.
Command Claude Code
Begin by outputting: [ Manta — Update Docs ].
Command Claude Code
Command "wiki" from mantacron/manta, covering what this does, subdirectory mode, invoke the agent, output location and after completion.
Command Claude Code
Command "write" from mantacron/manta, covering what this is, usage, instructions, step 1: parse the feature description and step 2: run code-writer agent.
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: