Trinity — Project Management agent for Claude Code
Implementation specialist.
How to install Trinity
Installs to ~/.claude/agents/vidit19sharma-neogents-trinity.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/vidit19sharma/neogents/HEAD/agents/trinity.md -o ~/.claude/agents/vidit19sharma-neogents-trinity.md Restart Claude Code, or start a new session, for it to be picked up.
What Trinity does
name: trinity description: Implementation specialist. Executes ONE well-defined task per invocation — exact files, exact changes, exact verification. Writes and runs tests for the behavior she changes. Follows the plan verbatim, verifies with evidence, never expands scope. The elite executor. model: sonnet tools: Read, Grep, Glob, Write, Edit, Bash
Trinity
You execute. One task, perfectly, with proof — then you stop. You receive a fully-specified task (from a plan artifact or the orc
Alternatives in Project Management
- Team Worker — Unified worker agent for team pipelines 530 ★
- Bolt Executor — Use this agent when there is a clear, well-scoped task to implement in code — a feature, fix, or refactor with 523 ★
- Grainulator — Autonomous research sprint subagent 87 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Fixme Execute Plan
Executes implementation plans step by step with maximum reliability. Follows plan exactly, verifies after ever
Coder Build
Implements approved plans and verifies with tests. Writes code, tests, and verification. Receives SESSION_ID a
Code Standards Reviewer
Use this agent when you need to verify that recently written or modified code follows the project's establishe
Echo Runner
Minimal mechanical runner — executes exactly the shell command given in the task prompt and relays its stdout
Implementation Lead
Execute approved specification tasks in small coherent slices, preserve repository conventions, add regression
Resection Implementation Reader
Cold reader of a project's actual current implementation — source, tests, config, and observable running behav
Related Skills
Test Locally
How to test pr-review changes locally — unit tests, end-to-end runs, and plugin installation verification.
Skillopt
Run the SkillOpt process-skill optimizer (offline, local). Default prints the exact bare-terminal command (saf
Issue Autofix Validate
Validate the autofix queue before a run — check each queued GitHub issue is defined well enough to fix automat