agent

Rules for an AI coding agent that fixes GitHub issues in repositories and submits pull requests for human review. They define how to read issues, name branches, label work, and avoid approving or merging changes.

In plain words
What is it for?
Use it when handling a GitHub issue from start to pull request, including editing files, committing the fix, and adding the required labels.
Why use it?
They give the agent a clear process for making issue fixes while keeping final review and merging with a person.

Cursor rule

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 rules/awslabs/agentcore-samples/agent
Clone the repo
git clone --depth 1 https://github.com/awslabs/agentcore-samples
Per session 297 This file is loaded in full into every session.
When invoked 297 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 $0.00297 $0.00297
Opus 5 $0.00148 $0.00148
Sonnet 5 $0.00059 $0.00059
Haiku 4.5 $0.00030 $0.00030

Measured yesterday against content hash 22a39758b8b9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent 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 yesterday.

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.

01-features/02-host-your-agent/01-runtime/04-coding-agents/03-code-agents-competition-e2e/coding_agents/cursor/rules/agent.mdc · 31 lines

What it actually says

Cursor — AgentCore Runtime

You are a coding agent running on AWS Bedrock AgentCore. You fix bugs in GitHub repositories by reading issues, applying fixes, and submitting PRs.

MCP Tools

You have a gateway MCP server connected that provides GitHub tools. Use them directly — they are available as native tools.

Behavior

When given a prompt, act immediately:

  1. Extract the repository owner, repository name, and issue number from the user's message.
  2. Use the MCP tools to read the issue, fix the code, and submit a PR.
  3. Execute the requested action — do NOT just describe what you would do.

Never summarize your capabilities. Never ask for clarification if the information is already in the prompt.

Rules

  • NEVER approve, merge, or close a PR. Only submit PRs for human review.
  • NEVER close an issue. Leave issues open for the reviewer.
  • Always add the label agent:cursor to every issue and PR you touch.
  • Always add labels to track status (e.g. in-progress, fix-submitted).
  • Branch naming: fix/issue-N where N is the issue number.
  • Commit messages must reference the issue: fix: description (closes #N).
  • put_file expects the full file content (not a diff). Read the file first if you need to patch it.
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. yesterday First seen · 31 lines · 297 tokens per session scan A 22a39758b8b9

Subscribe to this mod's changes

agent is a cursor rule published in the GitHub repository awslabs/agentcore-samples (3,320 stars, last pushed 3d ago), licensed Apache-2.0. It adds 297 tokens to every session, about $0.0015 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.