terminal-bench-science AGENTS.md

terminal-bench-science AGENTS.md is an instructions file for Codex, OpenCode from harbor-framework/terminal-bench-science. It costs 12,599 tokens per session, scanned A, original, Apache-2.0.

A guide for working on Terminal-Bench-Science, a benchmark that tests AI agents on real scientific research tasks.

In plain words
What is it for?
Use it when creating, reviewing, or submitting benchmark tasks and pull requests, or when running the documented validation commands.
Why use it?
It explains the repository layout, shared project template, review process, automated checks, and scoring rules for proposed tasks.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/harbor-framework/terminal-bench-science/agents-md
Clone the repo
git clone --depth 1 https://github.com/harbor-framework/terminal-bench-science

Made for: Codex, OpenCode.

Wrote 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.

agentmods badge for terminal-bench-science AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/harbor-framework/terminal-bench-science/agents-md.svg)](https://agentmods.dev/instructions/harbor-framework/terminal-bench-science/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/harbor-framework/terminal-bench-science/agents-md"><img src="https://agentmods.dev/badge/instructions/harbor-framework/terminal-bench-science/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 12,599 This file is loaded in full into every session.
When invoked 12,599 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.12599 $0.12599
Opus 5 $0.06300 $0.06300
Sonnet 5 $0.02520 $0.02520
Haiku 4.5 $0.01260 $0.01260

Measured 4d ago against content hash bddb647fbe2d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

terminal-bench-science AGENTS.md 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 4d 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.

AGENTS.md · 244 lines

How it starts

The opening of the file, as written. The whole thing — 244 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Terminal-Bench-Science

Agent benchmark built with Harbor. Shared CI, review bot, and task structure come from the benchmark-template repo via a template git remote. Terminal-Bench-Science is a natural science benchmark targeting real scientific research workflows.

Repo Structure

.
├── CONTRIBUTING.md                         # Contributor guide (task creation, submission)
├── REVIEWING.md                            # Reviewer guide (review order, DRI model, labels)
├── TASK_REVIEW_AUTOMATION.md               # CI pipeline docs (checks, commands, secrets)
├── TASK_IMPLEMENTATION_RUBRIC.toml         # Symlink/alias — see rubrics/task-implementation.toml
├── README.md                               # Public-facing overview with benchmark progress table
├── task-template.toml                      # Template for `harbor tasks init`
├── tasks/                                  # Benchmark tasks (each has instruction.md, task.toml, etc.)
├── rubrics/
│   ├── task-implementation.toml            # 39-criteria rubric for PR review (`harbor check`)
│   ├── task-proposal.md                    # LLM reviewer prompt for Discussions/Discord
│   ├── author-fit.md                       # LLM prompt for the separate author-aware pass (advisory author–task fit rating + COI flag)
│   └── trial-analysis.toml                 # 8-criteria trial-quality rubric for `harbor analyze` (per-trial, via -r)
├── ci_checks/
│   ├── check-*.sh                          # Static check scripts (canary, dockerfile, paths, etc.)
│   ├── check-*.py                          # Python checks (similarity, AI detection, task proposal link, rubric consistency)
│   ├── check-task-fields.sh             # task.toml schema validation
│   ├── rubric_review.py                    # LLM proposal rubric review + author-fit pass (--author-fit / --author-fit-only)
│   └── test-tasks/                         # Intentional-failure tasks for CI regression testing
├── .github/
│   ├── workflows/                          # CI workflows (see TASK_REVIEW_AUTOMATION.md)
│   ├── llm-config.yml                      # Default agents/models for /run and /cheat (harbor_run.* keys)
│   ├── reviewer-pool.yml                   # Reviewer pool: reviewers_by_field (domain) + reviewers_technical + reviewers_final (domain & technical reviewed in parallel)
│   ├── hack-trial-prompt.md                # Adversarial prompt for /cheat trials
│   └── pull_request_template.md            # PR template for task submissions
└── tools/
    ├── review-status/                      # PR review status report generator
    ├── rubric-regression/                  # Generates the fail-rubric-* review meta-task dataset (CI catch-rate gate)
    ├── batch-grader/                       # Batch grading tool
    └── task-readme/                        # Renders a task README.md from its task.toml (deterministic; no LLM)

Read the full file on GitHub · 244 lines

Changes

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.

  1. 4d ago First seen · 244 lines · 12,599 tokens per session scan A bddb647fbe2d

Subscribe to this mod's changes

terminal-bench-science AGENTS.md is an instructions file published in the GitHub repository harbor-framework/terminal-bench-science (379 stars, last pushed 5d ago), licensed Apache-2.0. It adds 12,599 tokens to every session, about $0.0630 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.

Related

Other instructions, from other repositories

wisp-science AGENTS.md

AGENTS.md instructions for xuzhougeng/wisp-science, covering agents.md, project orientation, repository layout, engineering rules and verification commands.

xuzhougeng/wisp-science · 2,195 tokens

MoleCode AGENTS.md

Instructions for AtomFlow-AI/MoleCode, covering agents.md — molecode, what this repo is, use the molecode skill, molecode-first rule and cli quick start.

AtomFlow-AI/MoleCode · 847 tokens

open-science CLAUDE.md

Claude Code instructions for ai4s-research/open-science, a project described as: Open Science Desktop — local-first, model-agnostic AI research workbench for macOS, Windows & Linux. Open-source Claude Science desktop alternative built on Tauri + MCP + agent skills.

ai4s-research/open-science · 3 tokens

preprint-fulltext copilot-instructions.md

Copilot instructions for genecell/preprint-fulltext, covering preprint-fulltext, when to use, why full text (for grounded reasoning), cli and python.

genecell/preprint-fulltext · 1,481 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens