autonomous-loop-operations

A skill for running an assistant in limited automatic cycles, where it checks queued events and ready tasks, decides whether to act, and stops under defined conditions.

In plain words
What is it for?
Configuring, testing, or running bounded background cycles that process queued tasks, scheduled wakeups, and allowed initiative.
Why use it?
It helps continuous assistants work without running indefinitely or acting outside their approval and safety rules.

Skill for Claude CodeCodex

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 skills/codeinfinity1/stram/autonomous-loop-operations
Any agent
npx skills add CodeInfinity1/Stram --skill autonomous-loop-operations
Clone the repo
git clone --depth 1 https://github.com/CodeInfinity1/Stram

Made for: Claude Code, Codex.

Per session 46 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 669 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 $0.00046 $0.00669
Opus 5 $0.00023 $0.00334
Sonnet 5 $0.00009 $0.00134
Haiku 4.5 $0.00005 $0.00067

Measured yesterday against content hash 782840dabef3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

autonomous-loop-operations 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.

skills/agent-core/autonomous-loop-operations/SKILL.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Autonomous Loop Operations

Purpose

Operate the assistant's autonomous loop safely: ingest stimuli, decide whether to act, execute bounded work, record evidence, and stop cleanly. This adapts external reference autonomy primitive ideas to Stram's automation daemon and cognition loop.

When To Use

Use when the user wants the agent to run continuously, process queued tasks, wake itself up, watch for ready work, or perform one bounded daemon tick.

Inputs And Evidence

  • autonomous-status or automation_daemon_status.
  • Queue entries, wakeups, triggers, recent cycles, and active commitments.
  • User-configured loop policy such as max cycles, idle behavior, and approval posture.
  • Current tool permissions and safety gates.

Tool Map

  • automation_daemon_status
  • automation_daemon_configure
  • automation_daemon_tick
  • cognitive_trigger_status
  • cognitive_trigger_evaluate
  • cognitive_priority_review
  • cognitive_state

Workflow

  1. Inspect daemon status and current queue before changing configuration.
  2. Confirm loop bounds: max cycles, idle stop condition, response mode, approval behavior, and whether initiative is allowed.
  3. Configure the daemon only when the user asks to persist loop settings.
  4. Run one tick for verification before suggesting a longer loop.
  5. Review tick output for actions taken, skipped work, approvals, and queued follow-ups.
  6. Record lessons or environment constraints when the loop exposes recurring blockers.

Safety And Boundaries

  • Never run unbounded loops without an explicit user request and clear stop conditions.
  • External-visible actions still require the channel/message/tool approval policy.
  • Do not enable initiative when the user asked only for passive monitoring.

Safety And Approval

  • Always set bounded max cycles, idle stop behavior, and initiative policy before running a loop.
  • Run a single verification tick before recommending continuous operation.
  • Keep high-risk, external-visible, destructive, or privileged actions paused for approval.
  • Stop or reduce scope on repeated failures, duplicate event processing, model/provider unavailability, or unclear queue state.

Read the full file on GitHub · 78 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 · 78 lines · 46 tokens per session scan A 782840dabef3

Subscribe to this mod's changes

autonomous-loop-operations is a skill published in the GitHub repository CodeInfinity1/Stram (10 stars, last pushed 22d ago), licensed MIT. It adds 46 tokens to every session and 669 once invoked, about $0.0002 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.