SKILLEMALL.ai
GitHub ↗ RU Sign in

CB terminal-bench-loop

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.

paperclipai/paperclip Agent Skills author: paperclipai MIT 1 file body ≈ 6 205 tokens Open the sourcegithub.com↗ analyzed 6 d ago

Run one Terminal-Bench task through a bounded Paperclip smoke/diagnosis/fix loop.

As a process B 72/100 · Nearly there — weak spots: result and completion, running it twice, progress reporting

AnalyzerSoftware developmenttype and topics are labelled automatically from the skill text
Runs in: Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) OpenClaw Goose OpenCode
JSON
Technical rating
C
87/100
safety, quality, tests
Safety 60%
99
Quality 40%
69
Run on models
none yet
Process rating
B
72/100
Nearly there
Progress reporting w 2
0
Running it twice w 4
30
Result and completion w 14
40
the three weakest of ten parameters · all ten

How to improve

  1. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 1

✓ No critical or high findings

Medium and low: 1
  • low Concealment en-hide-from-user SKILL.md:196
    Instruction to hide actions from the user (negated — the text forbids it)
    - **Hiding the chain.** Do not silently delete or hide failed iteration children, retracted proposals, or rejected confirmations. The audit trail is the loop's evidence.
    negated

Files scanned: 1. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning body-long SKILL.md body ≈ 6205 tokens (recommended < 5000); move details to references/

Process rating: all ten parameters 72/100

  • 0Progress reporting. Says nothing while it works
  • 30Running it twice. 16 mutating operations with no state check
  • 40Result and completion. Does not say what the result is
  • 60Tools and files. Uses tools (bash, node) that frontmatter does not declare
  • 70Inputs and preconditions. Inputs and preconditions are listed
  • 70Execution cost. Instruction body is 6205 tokens
  • 85Steps. 102 steps, 1 vague phrases
  • 100When it triggers. States when to use and when not to
  • 100Failures and branches. 3 branches, has a failure section
  • 100Consistency. Name and required fields are in place
  • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
  • low 11 top-level sections: this looks like several domains in one skill

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +5Description has no quoted example phrases that should trigger the skill
  • +4Description does not say when NOT to use the skill (false activations)
  • +3Output format is not stated: the model decides each time
  • -5TODO / placeholder text left in the skill
  • +1No license
  • +2Single-language instructions
  • +3Description length 215: enough signal without eating the budget
  • +4Structure: 22 headings
  • +3Step-by-step instructions: 102 items
  • +4Has examples (1 code blocks)

Quality base 70; lint remarks subtract, signals add up to 100. Result: 69.