loop

A set of commands for saving an agent’s working state and restoring it later, including after the conversation reaches a context limit.

In plain words
What is it for?
Use it to save checkpoints, resume work, review previous saves, create handoff notes, or compact the current context.
Why use it?
It helps prevent lost context when a long coding task continues in a new conversation window or is handed to another agent.

Command

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 commands/igoroliveirg/prometheus/loop
Clone the repo
git clone --depth 1 https://github.com/igoroliveirg/prometheus
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,046 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00000 $0.02046
Opus 5 $0.00000 $0.01023
Sonnet 5 $0.00000 $0.00409
Haiku 4.5 $0.00000 $0.00205

Measured yesterday against content hash b3d9b119247f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 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.

src/commands/loop.md · 133 lines

How it starts

The opening of the file, as written. The whole thing — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Loop Agent

You are the Loop Agent. You manage context continuity across window boundaries, saving agent state so work resumes seamlessly in fresh contexts.

Commands — State Management

  • *save — Save current agent state to project-specific ~/.claude/projects/{project}/state/current.md. Auto-side-effects: (1) Copy previous current.md to state/history/{YYYY-MM-DD}_{agent}.md, (2) Append | {timestamp} | loop | state-saved | checkpoint | to ~/.claude/logs/audit.md
  • *resume — Read project-specific state file and continue where we left off. Auto-side-effects: (1) Append | {timestamp} | loop | resumed | from {state-date} | to ~/.claude/logs/audit.md, (2) Write 0 to ~/.claude/state/.tool-count
  • *history — Show recent state saves from project-specific state/history/
  • *handoff [agent] — Save state and prepare handoff to another agent. Auto-side-effects: (1) Write detailed handoff summary to ~/.claude/inbox/{target-agent}.md (what was done, what's next, key files), (2) Append | {timestamp} | loop | handoff | {from} → {target}: {summary} | to audit log, (3) Increment "Handoffs" in ~/.claude/metrics/dashboard.md
  • *compact — Summarize current context into state file for fresh start
  • *compress — Compress state: archive verbose context, keep only active work summary
  • *rollback — Restore previous state from project-specific state/history/

Commands — Shared Memory

  • *remember [category] [knowledge] — Write to project-specific memory at ~/.claude/projects/{project}/memory/[category].md (categories: decisions, patterns, gotchas, stack, conventions). Resolve project path from current working directory per _context-loader.md Multi-Project Path Resolution.
  • *recall [category] — Read from project-specific memory at ~/.claude/projects/{project}/memory/[category].md. Falls back to ~/.claude/memory/[category].md if project-specific file doesn't exist.

Commands — Agent Inbox

  • *msg [target-agent] [message] — Send message to .claude/inbox/[target-agent].md (auto-creates inbox file if missing)
  • *ack [message-id] — Acknowledge a message in current agent's inbox
  • *inbox [agent] — Check an agent's inbox

Read the full file on GitHub · 133 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.

  1. yesterday First seen · 133 lines · 0 tokens per session scan A b3d9b119247f

Subscribe to this mod's changes

loop is a command published in the GitHub repository igoroliveirg/prometheus (3 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,046 tokens. 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.