What is floop AGENTS.md?

Contributor instructions for floop, a persistent-memory system for coding agents. It stores lessons from corrections and can load relevant working habits into later sessions.

Use it to save corrected behaviors, review active or stored memories, mark memories as helpful or wrong, and remove duplicates.

It helps an agent remember corrections across sessions instead of repeating the same mistake.

floop AGENTS.md — Instructions file by nvandessel

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 instructions/nvandessel/floop/agents-md

Clone the repo

git clone --depth 1 https://github.com/nvandessel/floop

Made for: Codex, OpenCode.

Per session 3,483 This file is loaded in full into every session.

When invoked 3,483 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	$0.03483	$0.03483
Opus 5	$0.01741	$0.01741
Sonnet 5	$0.00697	$0.00697
Haiku 4.5	$0.00348	$0.00348

Measured 2d ago against content hash a120bf808602, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

floop AGENTS.md 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 2d ago.

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.

AGENTS.md · 428 lines

How it starts

The opening of the file, as written. The whole thing — 428 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Floop - Agent Instructions

For floop contributors. If you're a user, see docs/integrations/ for setup guides.

Floop Integration (REQUIRED)

You have persistent memory via floop. Learned behaviors are loaded via MCP and, where supported, auto-injected via hooks.

When corrected, IMMEDIATELY capture it:

mcp__floop__floop_learn(right="what to do instead")

# With optional context about what went wrong:
mcp__floop__floop_learn(right="what to do instead", wrong="what you did (optional)")

# With explicit tags (optional, for pack filtering):
mcp__floop__floop_learn(right="what to do instead", tags=["topic", "category"])

Do NOT wait for permission. Capture learnings proactively. The hooks will also auto-detect corrections, but explicit capture is more reliable.

For non-Claude agents, see docs/integrations/agent-prompt-template.md.

Available MCP tools:

floop_active - See currently active behaviors for this context
floop_learn - Capture a correction (USE PROACTIVELY)
floop_feedback - Signal whether a behavior was helpful (confirmed) or contradicted (overridden)
floop_list - List all stored behaviors
floop_deduplicate - Merge duplicate behaviors

Codex Runtime Cadence (No Lifecycle Hooks)

In Codex environments, treat these as required pseudo-hooks:

Task start: Call floop_active with current file and task.
Context change (file/task/mode shift): Re-call floop_active.
Correction received: Immediately call floop_learn (no permission needed).
Behavior outcome: Call floop_feedback with confirmed or overridden.

If MCP is unavailable, use CLI fallback immediately:

floop active --file <path> --task <task> --json
floop learn --right "what to do instead" --wrong "what happened" --file <path>
floop list --json

Project Overview

floop is a CLI tool that enables AI agents to learn from corrections and maintain consistent behavior across sessions.

Read the full file on GitHub · 428 lines

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.

2d ago First seen · 428 lines · 3,483 tokens per session scan A a120bf808602

Subscribe to this mod's changes

floop AGENTS.md is an instructions file published in the GitHub repository nvandessel/floop (26 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 3,483 tokens to every session, about $0.0174 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.

Code review Learning and teaching Memory and context ai-agents behavior-learning claude-code cli golang mcp-server

floop AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Floop - Agent Instructions

Floop Integration (REQUIRED)

Codex Runtime Cadence (No Lifecycle Hooks)

Project Overview

What this file has done since we first saw it

Other instructions, from other repositories

n8n-skills CLAUDE.md

Tutorial_AwesomeModernCPP AGENTS.md

rag-code-mcp copilot-instructions.md

ab900 CLAUDE.md

agentconfig.org AGENTS.md

ken CLAUDE.md