agon

Plugin agon in AutoResearch-Factory/Agon. Commands read from the repo's manifest — 1 month ago.

Skills (5)

• aris

  Research-mindset catalog distilled from ARIS skills, for subagent Read on demand.

• arxiv-tools

  This skill should be used when the user asks to "search for papers", "get paper info", "download LaTeX source", "generate BibTeX citation", "find citing papers", or provides an arXiv ID or arXiv URL.

• novelty-check

  Use when checking novelty of a research idea or proposal. Extract core claims, do multi-source arxiv / Scholar / S2 search, grade each claim HIGH/MEDIUM/LOW. 基本就是 ARIS novelty-check skill, 但去掉了 cross-model verification 和 report schema.

• server-health

  Use when you need project-specific compute/server notes to pick a server or confirm a running task, if the project has filled its server manual or monitoring hooks.

• sibyl

  Research-persona catalog distilled from Sibyl prompts, for subagent Read on demand. 与 ARIS 互补 — ARIS 按阶段组织, Sibyl 按人格视角组织.

Slash commands (4)

• /agon:deep-lit-tick

  Run the scoped deep literature loop until relevant papers are saturated.

• /agon:experiment-tick

  Orchestrate the experiment scientist/screener/coder/auditor/reviewer loop for a workspace.

• /agon:idea-tick

  Create or continue ideas from a topic and run create/review/refine/deep-lit to dual-gate saturation.

• /agon:proposal-tick

  Run proposal refine/review/deep-lit loops in parallel until proposals are ready.

Subagents (12)

• deep-lit-reader

  Read one arXiv paper in depth, write its wiki note, and emit a deep-lit result JSON.

• env-validator

  Check whether the system environment satisfies prerequisites for running the research system.

• experiment-auditor

  Audit the latest experiment round's key conclusions, execution consistency, and scientific validity.

• experiment-coder

  Implement, deploy, monitor, sync, and debug experiments from the STATE.md Runs table.

• experiment-reviewer

  Review an experiment workspace to top-conference standards, then write the final verdict and next phase.

• experiment-scientist

  Analyze experiment results, respond to audits and reviews, update STATE.md, and plan the next experiment round.

• experiment-screener

  Screen experiment plans before implementation, blocking unnecessary scale and meaningless gates.

• idea-creator

  Research a topic landscape, generate and select research ideas, and save them under ideas/.

• idea-refiner

  Improve an existing idea using reviewer feedback and write the next idea version.

• idea-reviewer

  Review an idea for novelty and quality, write the review, and update its ideas.xml score.

• proposal-refiner

  Write a proposal from an idea, or revise the latest proposal based on a review.

• proposal-reviewer

  Review a proposal across multiple dimensions, write the review, and update its proposals.xml score.

Hooks (8)

• PreToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/deny-durable-cron.py"

• PostToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/settings-change-notify.py" "${CLAUDE_PLUGIN_ROOT}/.settings.toml"

• PostToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/backend-limit-retry.py"

• PostToolUseFailure

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/backend-limit-retry.py"

• StopFailure

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/backend-limit-retry.py"

• UserPromptSubmit

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/backend-limit-retry.py"

• SessionStart

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/backend-limit-retry.py"

• SessionEnd

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/settings-change-notify.py" --cleanup