AutoGPT: Reproduce A High-Severity Failure & Write a CRE Rule [Multiple Winners] [Submit by August 31 11:59 pm ET]

[Lyndon-prequel]

Exclusive

⭐ star these repos and be the first to see new bounties

• https://github.com/prequel-dev/preq
• https://github.com/prequel-dev/cre

Challenge

AutoGPT is a pioneering autonomous agent framework that chains LLM reasoning with real-world actions — but subtle prompt-handling flaws, resource mismanagement, or runtime crashes can silently derail agent execution, trigger runaway task loops, or cause unexpected failures across integrated systems.

Reproduce a high-severity AutoGPT failure on a recent release and write a CRE that reliably flags this issue in real-world production environments.

Prizes

🏆 Technical Champion - Awarded to the highest quality PR

• All on-time PRs will be reviewed for
  • difficulty
  • quality
  • completeness
• The winning PR will be merged, you will receive the full bounty, and get featured on the Prequel blog.

⭐ Community Champion - Awarded to the submission most loved by the community

• To participate in this category, you must
  • Share your PR on X or linkedin
    • tag:
      • @prequel-dev and @auto_gpt (x)
      • @prequel-dev and @autogptofficial (linkedin)
    • describe the problem you are detecting
  • Embed a link to your social posts in your PR comment
  • Follow prequel on X @prequel_dev and LinkedIn [prequel-dev] (https://www.linkedin.com/company/prequel-dev/)
• The winning PR will be merged and you will receive a $50 bounty.

🙌 All other merged PRs will receive a $20 bounty. (Max 10)

✅ You must:

1. Reproduce the failure scenario
2. Share a minimal, working reproduction (e.g., Helm, Docker Compose, etc.)
3. Write a unique CRE-format detection rule for preq
4. Install preq and run the rule

📦 Deliverables:

• Reproduction setup and clear README
  • create a private repo and the reviewers (@tonymeehan @Lyndon-prequel)
• Link to your unique rule in the CRE playground
• PR containing
  • the new CRE rule
  • example logs in test.log
  • updated tags.yaml and category.yaml files
• Short demo video of
  • your reproduction
  • running the rule. i.e cat test.log | preq -r your_rule.yaml -d

💰 /bounty $200

[algora-pbc]

💎 $200 bounty • Prequel

Steps to solve:

1. Star the repos: https://github.com/prequel-dev/preq + https://github.com/prequel-dev/cre
2. Start working: Comment /attempt #129 with your implementation plan.

• If this is a generic bounty, we suggest you specify the specific issue you plan to work. There are cases where we will award multiple bounties.

3. Submit work: Create a pull request including

• closes #129 to attach to the issue
• /claim #129 in the PR body to claim the bounty

4. Receive payment: 100% of the bounty is received 2-5 days post-reward. Make sure you are eligible for payouts

❗ Important guidelines:

• To claim a bounty, you need to provide a short demo video of your changes in your pull request
• If anything is unclear, ask for clarification before starting as this will help avoid potential rework
• Low quality AI PRs will not receive review and will be closed
• Do not ask to be assigned unless you've contributed before

Thank you for contributing to prequel-dev/cre!

Attempt	Started (UTC)	Solution	Actions
🟢 @MAVRICK-1	Aug 27, 2025, 04:04:59 AM	#149	Reward
🟢 @Harsh9485	Aug 27, 2025, 05:32:24 AM	WIP
🟢 @RaghavArora14	Aug 28, 2025, 03:15:14 PM	WIP
🟢 @himayday	Aug 29, 2025, 10:31:20 AM	#140	Reward
🟢 @rvhost	Aug 30, 2025, 08:05:24 AM	WIP
🟢 @dhvll	Aug 31, 2025, 10:34:57 AM	#145	Reward
🟢 @piyzard	Aug 31, 2025, 12:41:32 PM	#147	Reward
🟢 @Sahelisaha04	Aug 31, 2025, 06:48:39 PM	#151	Reward

[MAVRICK-1]

/attempt #129

[Harsh9485]

/attempt #129

[RaghavArora14]

/attempt #129

[yashx007]

/attempt #128

[himayday]

/attempt #128

[piyzard]

/attempt #128

[yashx007]

/attempt #141

[rvhost]

/attempt #129