Skip to content
Get packs

Fix Failed Paperclip Runs

Diagnoses and repairs failed or stuck Paperclip tasks, then delivers a tested PR

Source Download skill
Repository
paperclipai/paperclip
Stars
110,078
Licence
MIT
Files
2 files

What it does

When a Paperclip task or agent run fails, gets stuck, runs slowly, or is wrongly blocked, this skill tells the AI agent to trace the root cause from logs and records, write a test that reproduces the bug and fails before any fix, make a targeted repair, check side effects, and deliver a pull request whose checks pass. It is a repair playbook for maintaining the Paperclip project itself, not a general tool for end users. The agent is told that a diagnosis or a passing unit test is not enough; the goal is a verified workflow and a green PR, and that merging or deploying is not authorized by the skill alone.

Great for

  • Maintainers and contributors of the Paperclip repository who need to investigate a failed or stuck run
  • Developers who have a task or run URL and want an agent to trace and propose a tested fix

Skip it if

  • Users who only want to run Paperclip tasks: the skill is written for working on the Paperclip codebase and assumes access to instance logs and repository tooling
  • Production incidents where the agent might push or deploy changes: the skill says it does not itself authorize merging or deploying, so human review of any rollout is still needed

Try asking

  • “Investigate why this Paperclip run failed and write a regression test that reproduces it”
  • “This task has been stuck in waiting for hours; find out whether it is a real block or a routing bug”
  • “Fix the slow run reported in this failure report and open a PR with passing checks”
Step-by-step guide: how to use Fix Failed Paperclip RunsInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the MIT licence and an attribution note.

    fix-failed-paperclip-run.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip fix-failed-paperclip-run.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip fix-failed-paperclip-run.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Give the agent a concrete task or run URL plus the observed symptom and expected result, so it can establish the finish line
  • Ask for diagnosis-only if you want findings without code changes, since the skill honors that request
  • Review the PR yourself before merging or rolling out; the skill keeps merge and deploy decisions with you

What's inside

  • references/failure-patterns.md
  • SKILL.md

Redistributed unmodified from paperclipai/paperclip under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.