Agent Claude Code
Utility agent for pulling benchmark data, evaluating cost vs performance, and recommending which models to use for which agents. Use on-demand for model assignment reviews.
not rated 4 3mo ago A 35 tokens
Agent Claude Code
Multi-causal analysis with cross-module tracing and architecture-level diagnosis.
not rated 4 3mo ago A 18 tokens
jg-debugger
03Agent Claude Code
Failure classifier and diagnostician. Analyzes test failures and produces fix instructions.
not rated 4 3mo ago A 20 tokens
jg-git
04Agent Claude Code
Handles branching, conventional commits, and PR creation. Use after reviewer passes. Does not merge.
not rated 4 3mo ago A 23 tokens
jg-planner
05Agent Claude Code
Coordinates the implementation pipeline with tiered agent routing. Classifies task complexity and dispatches to appropriate agent tiers.
not rated 4 3mo ago A 26 tokens
Agent Claude Code
Scope check and lint-level review for trivial changes.
not rated 4 3mo ago A 15 tokens
Agent Claude Code
Deep review with architecture analysis, security implications, and backward compatibility checks.
not rated 4 3mo ago A 19 tokens
jg-reviewer
08Agent Claude Code
Quality gate before commit. Reviews diff for scope creep, overengineering, and unnecessary complexity. Use after tests pass.
not rated 4 3mo ago A 27 tokens
Agent Claude Code
Decomposes complex issues with dependency graphs, risk analysis, and rollback strategies.
not rated 4 3mo ago A 21 tokens
Agent Claude Code
Decomposes issues into structured implementation plans with ordered steps and acceptance criteria mapping. Use when breaking down a complex issue into actionable tasks.
not rated 4 3mo ago A 31 tokens
Agent Claude Code
Phase 1 verification only (lint, typecheck, unit tests). For trivial changes.
not rated 4 3mo ago A 23 tokens
jg-tester
12Agent Claude Code
Two-phase verification gate. Phase 1: lint, typecheck, unit tests. Phase 2: integration and E2E.
not rated 4 3mo ago A 31 tokens
Agent Claude Code
Implements single-file edits and simple changes. Escalates if task exceeds scope.
not rated 4 3mo ago A 21 tokens
jg-worker
14Agent Claude Code
Implements code and tests per plan; reports completion to the planner.
not rated 4 3mo ago A 17 tokens
Cursor rule Cursor
Commit and PR conventions.
not rated 4 3mo ago A 4 tokens
Cursor rule Cursor
Issue-as-source-of-truth and start/completion workflow.
not rated 4 3mo ago A 11 tokens
Cursor rule Cursor
Delegate multi-step work to jg-planner first.
not rated 4 3mo ago A 574 tokens
jg-pr-review
18Cursor rule Cursor
PR review categories and decision rules.
not rated 4 3mo ago A 6 tokens
Cursor rule Cursor
Complexity classification and tiered agent routing.
not rated 4 3mo ago A 398 tokens
Skill Claude CodeCodex
Benchmark collection and evaluation workflow for agent model assignment reviews. Use when pulling benchmarks, evaluating cost/performance, or deciding which models to use for which agents.
not rated 4 3mo ago A 37 tokens
Skill Claude CodeCodex
Read/write layout for pipeline artifacts in .pipeline/. Use when any jg- agent reads upstream artifacts or writes its output.
not rated 4 3mo ago A 33 tokens
rule
22Cursor rule Cursor
Cursor rule "rule" from JahnelGroup/multi-agents, covering jg, when to apply and exempt (optional).
not rated 4 3mo ago A 6 tokens
explain-code
23Skill Claude CodeCodex
Explains code with visual diagrams and analogies. Use when explaining how code works, teaching about a codebase, or when the user asks, "how does this work?".
not rated 2 7mo ago A 39 tokens
Skill Claude Code
Inspect git history and append work summary to the daily work log.
not rated 2 7mo ago A 17 tokens
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: