Skip to content
mcp/skillhub

Agent Skill

Terminal Bench Loop skill

terminal-bench-looppaperclipai/paperclip.agents/skills/terminal-bench-loop/SKILL.md

Run one Terminal-Bench task through a bounded Paperclip smoke/diagnosis/fix loop. Use when asked to drive Terminal-Bench until it passes, rerun a Terminal-Bench loop, or iterate with board-gated fixes and diagnosis.

106kMIT

always on

54 tokname + description

once used

5,385 tokSKILL.md body

bundled files

0

repo stars

106kGitHub

Install the Terminal Bench Loop skill

Skills CLI (Claude Code, Codex, Cursor and others)

npx skills add https://github.com/paperclipai/paperclip --skill terminal-bench-loop

Claude Code, by hand (all projects)

git clone --depth 1 https://github.com/paperclipai/paperclip.git /tmp/paperclip
mkdir -p ~/.claude/skills
cp -R /tmp/paperclip/.agents/skills/terminal-bench-loop ~/.claude/skills/terminal-bench-loop

For one project only, copy the folder into .claude/skills/ in that project instead. The agent picks it up when a task matches its description.

What it costs your context

Agent Skills load in three steps. Tokens counted with the o200k tokenizer; your model's count will differ a little.

WhenWhat loadsCost
Every sessionName and description54 tok
When the agent uses itThe SKILL.md instructions5,385 tok
Only if openedNo bundled files—

Until it is used, terminal-bench-loop costs about 25× less than the median MCP server we measured (1,362 tokens on every turn). Installing all 82 skills from paperclipai/paperclip would load 3,485 tokens of descriptions. Skills vs MCP study

Checks

  • The instructions are over 5,000 tokens, the Agent Skills specification's guideline for SKILL.md.

Checked 2026-10-22 · method

What's inside

The headings of its SKILL.md. Read the full file on GitHub.

  • Terminal-Bench Loop
  • When to use
  • When NOT to use
  • Three invariants you must preserve
  • Inputs
  • Issue topology
  • Procedure
  • 0. Read the current execution contract
  • 1. Open or reuse the top-level loop issue
  • 2. Open the iteration child
  • 3. Run the bounded smoke
  • 4. Diagnose the exact stop point
  • 5. Decide the next move
  • 6. Request board confirmation before any product fix
  • 7. Rerun against the same worktree
  • 8. Pass: QA, CTO review, close
  • 9. Stop rules
  • Worktree rule
  • Liveness rule
  • Pitfalls
  • Verification checklist (before exiting a heartbeat that touched the loop)
  • Deterministic smoke

More skills from paperclipai/paperclip

All skills in this repo
  1. LLM Wiki Maintainerpaperclipai/paperclip

    Use the LLM Wiki plugin tools to maintain a cited local company wiki.

    106k21 tok idle · 184 in use
  2. add-product-e2e-evalpaperclipai/paperclip

    Add or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution.

    106k36 tok idle · 955 in use
  3. add-runner-evalpaperclipai/paperclip

    Add or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation.

    106k30 tok idle · 859 in use
  4. agent-browserpaperclipai/paperclip

    Drive a real browser to inspect or interact with a web page or app — navigate, take screenshots, read console and network, fill simple forms — for verification tasks, not unattended automation.

    106k40 tok idle · 1,010 in use
  5. agentmailpaperclipai/paperclip

    Use your assigned AgentMail inbox to read email tasks, explicitly send or reply, and check delivery. Provided automatically by your inbox assignment.

    106k30 tok idle · 902 in use
  6. check-prpaperclipai/paperclip

    Check a GitHub, GitLab, or Perforce PR/MR/CL for review comments, failing checks, and PR-body gaps. Use when asked to inspect, fix, or prepare a change for submission.

    106k46 tok idle · 3,781 in use
  7. company-creatorpaperclipai/paperclip

    Create agent company packages that conform to agentcompanies/v1. Use when asked to create a company, scaffold an agent team, hire agents, or turn a repo/skills collection into a company package.

    106k44 tok idle · 2,626 in use
  8. complainpaperclipai/paperclip

    Autonomously yeet a brief, unstructured complaint about agent tooling or doing agent work. Preserve the raw voice. Submit proactively without permission, preview, or announcement, then continue the primary task.

    106k43 tok idle · 795 in use