commet

24 mods across 2 repositories, 1 stars between them.

argus

01

commet/Argus

Plugin Claude Code

Plugin marketplace listing 1 plugin: argus.

1 2d ago A tokens not measured

SessionStart

02

commet/Argus

Hook Claude Code

Runs when a session starts, executing session-guard.sh via bash. From commet/Argus.

1 2d ago A tokens not measured

settings

03

commet/Argus

Settings file Claude Code

Agent settings declaring 1 hook event (SessionStart) and 7 allowed tools.

1 2d ago A tokens not measured

Argus CLAUDE.md

04

commet/Argus

Instructions file

A set of Claude Code instructions for Argus, including its repository map, commands, and product guidance. It also explains how this file fits with the project's six source-of-truth documents.

1 2d ago A 6,165 tokens

argus

05

commet/Argus

Plugin Claude Code

Turns a stuck decision into one observable next move, preserves the user-adopted record, and returns when reality can shape the next judgment.

1 2d ago A tokens not measured

argus-decision

06

commet/Argus

MCP server Claude CodeCodexCursor

MCP server "argus-decision" as configured in commet/Argus. Runs ${CLAUDE_PLUGIN_ROOT}/scripts/mcp-launch.js with node. Needs 1 environment variable to run.

1 2d ago A tokens not measured

domain-reviewer

07

commet/Argus

Agent

Reviews the target using the domain and implementation constraints visible in the project. Use as the primary specialist in deep review.

1 2d ago A 27 tokens

evidence-reviewer

08

commet/Argus

Agent

Independently checks the evidence behind a consequential judgment. Use only in deep review when factual or external claims are load-bearing.

1 2d ago A 30 tokens

risk-reviewer

09

commet/Argus

Agent

Challenges failure modes for critical or hard-to-reverse judgments. Use only when the decision is high-stakes or irreversible.

1 2d ago A 28 tokens

synthesizer

10

commet/Argus

Agent

Produces one concise synthesis or targeted revision from bounded reviewer findings while preserving disagreements and provenance.

1 2d ago A 22 tokens

UserPromptSubmit

11

commet/Argus

Hook

Runs when you submit a prompt, before the agent sees it, executing anchor-signal.js, ambient-nudge.js and sense-signal.js via node (3 commands). From commet/Argus.

1 2d ago A tokens not measured

PreToolUse

12

commet/Argus

Hook

Runs before the agent uses a tool for Bash, apply_migration, deploy_edge_function, delete_branch, pause_project and restore_project tool calls, executing keel-signal.js via node (2 commands). From commet/Argus.

1 2d ago A tokens not measured

PostToolUse

13

commet/Argus

Hook

Runs after a tool call finishes for Bash, Write and Edit tool calls, executing commit-signal.js and validate-gates.mjs via node with --hook (2 commands). From commet/Argus.

1 2d ago A tokens not measured

Stop

14

commet/Argus

Hook

Runs when the agent finishes a response, executing wake-signal.js and validate-gates.mjs via node with --latest and --warn (2 commands). From commet/Argus.

1 2d ago A tokens not measured

check

15

commet/Argus

Skill Claude CodeCodex

Check what is due, record what actually happened, save a candidate for later checking, or inspect a decision's premises. Use for "지금 확인할 것 있어?", "정산하자", "how did that bet go?", a candidate id, or premises. Invoked as /argus:check.

1 2d ago A 67 tokens

help

16

commet/Argus

Skill Claude CodeCodex

Explain Argus and route the user to the right command. Use when the user asks what Argus is, what commands exist, how to start, which command fits their situation, or where session files live. Read-only, no LLM pipeline, no session mutation. Invoked as /argus:help.

1 2d ago A 65 tokens

history

17

commet/Argus

Skill Claude CodeCodex

Read the decision record, inspect one decision's versions, keep a user-authored principle, or recover candidates from past chats. Never score or rank outcomes. Invoked as /argus:history.

1 2d ago A 42 tokens

loop

18

commet/Argus

Skill Claude CodeCodex

Turn a consequential, stuck decision into one observable next move, preserve only what the user explicitly adopts, and return when reality can answer. Use when the user is choosing between paths, committing to a consequential course, designing a test, or revisiting a past decision. Works across Agent Plugins hosts…

1 2d ago A 77 tokens

review

19

commet/Argus

Skill Claude CodeCodex

Deep pressure-test of a decision or artifact (PR, file, branch, document) — the full Argus reviewer pipeline. Sharpens the question, deploys reviewer agents on the real artifact in parallel, splits claims into supported/challenged/human-required, optionally runs a stakeholder pass, and returns one consolidated card.…

1 2d ago A 80 tokens

settings

20

commet/Argus

Skill Claude CodeCodex

Configure Argus, connect the webapp through browser approval, sync records, or inspect pairing state. Never ask for or print a token in normal use. Invoked as /argus:settings.

1 2d ago A 42 tokens

find-skills

21

commet/Argus

Skill Claude CodeCodex

Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

1 2d ago A 67 tokens

commet/Argus

Skill Claude CodeCodex

Postgres best practices maintained by Supabase, for Postgres running anywhere. Load this skill BEFORE writing or changing anything that lives in a Postgres database: creating or altering tables and columns (including choosing column types), schema design, migrations and declarative schema files, RLS policies and the…

1 2d ago A 182 tokens

commet/Argus

Skill Claude CodeCodex

React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance…

1 2d ago A 67 tokens

argus-decision-mcp

24

commet/Argus

MCP server Claude CodeCodexCursor

Preserve a user-authorized judgment and append later answers. No model verdict or human score. Runs locally from the argus-decision-mcp npm package. Needs 3 environment variables to run.

1 2d ago A tokens not measured