Agent Cursor
GOAL: One document per domain. Minimum tokens for maximum clarity.
not rated 88 1mo ago A 21 tokens
Demonstrate
26Agent
Agent for demonstrating VS Code features.
not rated 77 4mo ago A 10 tokens
AGENTS
27Agent
A set of rules for the model-adapter part of an inference gateway. An inference gateway connects applications to different AI model backends, while an adapter gives those backends one consistent interface.
not rated 76
+1 1mo ago A 0 tokens
original Apache-2.0
npc-mind
28Agent Claude Code
A private-mind agent for one named character at a time. It reacts only from that character’s personal point of view and must be given the character’s ID.
not rated 72 9d ago A 101 tokens
Agent
The opinionated shape-maker. Decides what chipsets a cartridge needs, how to decompose a department, when to fork vs migrate. Routes work to the other forge agents.
not rated 69 1mo ago A 40 tokens
Agent
Review changes for security vulnerabilities and trust-boundary issues.
not rated 68
+1 2d ago A 14 tokens
original MIT
analyst
31Agent Claude Code
Use this agent when performing exploratory data analysis, creating visualizations, running statistical tests, analyzing experiment results, or generating reports. For example: profiling a new dataset, creating distribution plots, running hypothesis tests on A/B experiment data, comparing model metrics across…
not rated 67 6mo ago A 73 tokens
original MIT
NOTICE
32Agent
Source: https://github.com/msitarzewski/agency-agents Pinned commit: 64eee9f8e04f69b04e78e150d771a443c64720be License verified: MIT, confirmed at the pinned commit on 2026-07-28 (GitHub license API, spdxid: MIT). Because the import was commit-pinned, a later upstream relicense does not affect the grant that applied at…
not rated 57
+1 2d ago A 0 tokens
original MIT
Agent Claude Code
Thin Claude adapter for experimental-attacker; canonical brief is authoritative.
not rated 50
+9 today A 17 tokens
MIT
Agent
You are a coding agent from the Hello World extension.
not rated 48
+2 4mo ago A 0 tokens
copy · 100% Apache-2.0
designer
35Agent
DESIGN SEAT for member-surface waves. Judges rendered screens and customer-visible copy against product dialect, decision economics (one decision, one gesture, evidence collapsed), copy humanity, card excerpt discipline, the ICP lens (trades owner on a phone in a truck) and the sauce (action-not-dashboards). Records a…
not rated 45
+2 yesterday A 93 tokens
original MIT
Agent
Budget-aware task routing using Qualixar OS cost engine. Analyzes task complexity, selects the cheapest model that meets quality requirements, and tracks spend against budget. Uses POMDP routing for optimal cost/quality tradeoff.
not rated 41 3mo ago A 51 tokens
Agent
You are the team lead responsible for orchestrating work across specialized agents. You create plans, delegate tasks, and ensure quality through validation.
not rated 38 4mo ago A 0 tokens
original MIT
amp
38Agent
Feature and research work are NOT complete until you run these commands yourself.
not rated 25
+1 4d ago A 0 tokens
original Apache-2.0
providers
39Agent
A coding agent runs against a provider account: a login, on a plan, that may or may not be metered. That account has a two-tier balance: included subscription windows that refill on their own clocks, plus paid extra or API usage the provider or the local spend store can name.
not rated 23
changed today A 0 tokens
original MIT
middleware
40Agent Codex
How to add middleware to an agent.
not rated 23 4mo ago A 7 tokens
original MIT
developer
41Agent Claude Code
Use this agent for software architecture design, feature implementation, debugging, refactoring, and test automation. Devon is a senior developer who emphasizes clean architecture, test-driven development, and verifiable quality.
not rated 22 5mo ago A 41 tokens
reviewer
42Agent Claude Code
Comprehensive code reviewer. Use when asked to review a PR, diff, branch, or specific files/folders. Spawns 7 specialized sub-reviewers in parallel, then synthesizes findings into a prioritized verdict.
not rated 22 3d ago A 46 tokens
original MIT
claude-cli
43Agent
MARKER — read-only Claude Pro lane. The director must supply prompt bytes on stdin to claude -p --model opus --permission-mode plan, NOT call workflow agent() with this agentType. See pi/SKILL.md "Claude routing rule (hard rule)".
not rated 21
+2 22d ago A 61 tokens
Agent Claude Code
Reviews a RondoFlow code change against this project's specific conventions and security rules (childprocess spawn safety, the { success, error } API envelope, per-user ownership/IDOR, Zod boundary validation, immutability, i18n parity, file/function size limits, Claude-auth handling). Use after writing or before…
not rated 20 yesterday A 112 tokens
original MIT
Test Worker
45Agent Claude Code
Autonomous agent specialized in test development. Creates comprehensive test suites.
not rated 20 yesterday A 15 tokens
AGPL-3.0
Agent Claude Code
Scans the Backlog column of teamctl's GitHub Project board and triages each item for hygiene — already-shipped, duplicate, or stale — against the repo and the linked issue. Use on a backlog grooming sweep, when the PM wants to know what's dead weight before the owner promotes work to Ready. Returns a proposal list…
not rated 20 1mo ago A 98 tokens
original MIT
Architect
47Agent Claude Code
Design decisions, ADRs, technical specifications.
not rated 19 1mo ago A 10 tokens
original MIT
Agent Claude Code
Create foundational templates, boilerplate code, and starter configurations for new projects, components, or features. Use for scaffolding a new component, API endpoint, model, config, or test suite from scratch.
not rated 18 yesterday A 45 tokens
original MIT
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: