What is agentic-loop?

A set of development rules for a small Go bridge that passes standard input messages to a pREST HTTP service. It defines a repeatable loop: understand, locate, plan, implement, validate, and review.

Use it when changing the bridge to locate the right packages, inject dependencies, handle configuration and errors, format code, and run validation checks.

It keeps changes focused on transport and configuration, while protecting the message protocol, error handling, and project structure.

agentic-loop — Cursor rule by prest

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/prest/prest-mcp-adapter/agentic-loop

Clone the repo

git clone --depth 1 https://github.com/prest/prest-mcp-adapter

Made for: Cursor.

Per session 508 This file is loaded in full into every session.

When invoked 508 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.

Model	Per session	Once invoked
Fable 5	$0.00508	$0.00508
Opus 5	$0.00254	$0.00254
Sonnet 5	$0.00102	$0.00102
Haiku 4.5	$0.00051	$0.00051

Measured today against content hash 79cd2b8126e8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agentic-loop 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 today.

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.

.cursor/rules/agentic-loop.mdc · 71 lines

What it actually says

prest-mcp-adapter Agentic Development Loop

Follow this loop on every development task. For conventions, see @.github/copilot-instructions.md.

The Loop

Understand → Locate → Plan → Implement → Validate → Self-review → (iterate)

1. Understand

Restate the goal: this repo is a dumb stdio→HTTP bridge to pREST /_mcp.
Check go.mod for the Go version before choosing APIs.
Do not add schema/SQL/tool business logic unless explicitly requested.

2. Locate

Path	Role
`cmd/prest-mcp/`	Entrypoint
`internal/config/`	Env loading
`internal/transport/`	Stdio NDJSON + HTTP POST
`internal/proxy/`	Forward + JSON-RPC error synthesis
`internal/logging/`	slog → stderr

3. Plan

Smallest diff that solves the task.
Keep stdout protocol-only; logs on stderr.
Fail closed on missing/invalid PREST_MCP_URL.

4. Implement

Area	Rule
Scope	Tiny transport bridge; no pREST domain logic
State	Inject deps via constructors; no package-level mutable app state
Errors	Wrap with `%w`; use `slog` on stderr
Formatting	`gofmt`

5. Validate

go build ./...
go vet ./...
gofmt -l .
make test       # includes overall coverage >= 80%

6. Self-review

Diff is minimal
No logs on stdout
Config fail-closed for required URL
New behavior has unit tests (httptest / buffers)
Overall statement coverage stays ≥80%
Validation commands passed

Definition of Done

go build ./... and go vet ./... pass
make test passes (-race -timeout 30s + overall coverage ≥80%)
Self-review checklist is green
No scope creep

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.

today First seen · 71 lines · 508 tokens per session scan A 79cd2b8126e8

Subscribe to this mod's changes

agentic-loop is a cursor rule published in the GitHub repository prest/prest-mcp-adapter (0 stars, last pushed 1mo ago), licensed MIT. It adds 508 tokens to every session, about $0.0025 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 2026-08-31.

Code review Databases Planning adapter ai-agents claude cursor go

sample_cursor_rule

React component best practices.

ZanzyTHEbar/cursor-rules · 88 tokens

cursor

You are working on the checkout service. Preserve transaction integrity and auditability.

mrwogu/promptscript · 53 tokens

beanstalk-deploy

Robust deployment patterns for Elastic Beanstalk with GitHub Actions, Pulumi, and edge case handling.

pr-pm/prpm · 0 tokens

creating-kiro-agents

Kiro agent configuration patterns, JSON structure, tool permissions, and security best practices for creating specialized AI development assistants.