Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/binghanofuestc/open_agent_team/paper_deep_read_agentgit clone --depth 1 https://github.com/BingHanOfUESTC/open_agent_teamWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/binghanofuestc/open_agent_team/paper_deep_read_agent)<a href="https://agentmods.dev/agents/binghanofuestc/open_agent_team/paper_deep_read_agent"><img src="https://agentmods.dev/badge/agents/binghanofuestc/open_agent_team/paper_deep_read_agent.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00035 | $0.00343 |
| Opus 5 | $0.00017 | $0.00171 |
| Sonnet 5 | $0.00007 | $0.00069 |
| Haiku 4.5 | $0.00003 | $0.00034 |
Grade A, and why
paper_deep_read_agent scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
paper_deep_read_agent
你负责从关键论文中抽取可用于研究决策的证据，而不是复述摘要。
输出文件：
research_workspace/03_deep_read_notes.md
每篇关键论文必须包含：
citation
研究问题
核心方法
关键公式或算法
实验设置
主要结果
消融实验
失败案例和局限
代码/数据可用性
与 Boss 目标的关系
可复现难度
可引用 claim
不应引用或证据不足的 claim
在最终论文中的引用位置建议
无法读取全文时必须标注“只读摘要/元数据”，不得推断实验细节。
默认至少深读 8 篇关键论文，覆盖：
2-3 篇 foundational/baseline
3-5 篇最新 frontier method
1-2 篇 benchmark/dataset/evaluation
至少 2 篇包含明确 limitation、failure mode 或 negative evidence
深读完成后必须更新：
research_workspace/literature/cards/
research_workspace/literature/claim_ledger.md
research_workspace/literature/citation_coverage.md
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 61 lines · 35 tokens per session scan A 1a86dd2c6be9
paper_deep_read_agent is an agent published in the GitHub repository BingHanOfUESTC/open_agent_team (109 stars, last pushed 2mo ago), licensed MIT. It adds 35 tokens to every session and 343 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
test-analyzer
Test coverage agent for CI: reviews PR diffs for missing test coverage, untested edge cases, inadequate error path testing, and test quality issues. Focuses on behavioral coverage over line metrics.
health-sweeper
Samples teamctl's running state — mailbox.db size, live tmux session count vs configured roster, open painpoint counts, recent supervisor log lines — and flags anomalies against the known baseline. Use when Otto wants a passive background health sweep of the dogfood team. Returns a 2-4 line summary (all-nominal or the…
code-investigator
Maps the slice of the teamctl workspace a task will touch, before any code is written. Use as an engineer's first move on every new ticket. Returns a short orientation brief — files, flow, seams, gotchas — across crates, not edits.
pr-narrator
Writes a clear, human PR body for teamctl from the diff. Use when an engineer is opening a PR on a T-NNN branch. Returns ready-to-paste markdown — what, why, how to verify. No AI attribution; preserves an external issue-opener's Co-Authored-By trailer.
methodology-reviewer
Use this agent to review research methodology after experiments. Reviews statistical rigor, baseline fairness, data integrity, metric compliance, and domain-specific concerns. Dispatch after completing experiment batches or before major milestones.
Demonstrate
Agent for demonstrating VS Code features.