agentic-loop

A workflow that lets a developer approve a build plan once, then hands implementation, testing, review, and limited fixes to separate agents.

In plain words
What is it for?
Use it for clearly specified coding tasks that can run in separate slices, with an independent review before completion.
Why use it?
It removes the need to supervise each step of a well-defined task. The process stops when it cannot continue without guessing.

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/asteasolutions/ai-toolkit/agentic-loop
Any agent
npx skills add asteasolutions/ai-toolkit --skill agentic-loop
Clone the repo
git clone --depth 1 https://github.com/asteasolutions/ai-toolkit

Made for: Claude Code, Codex.

Per session 38 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 923 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 $0.00038 $0.00923
Opus 5 $0.00019 $0.00462
Sonnet 5 $0.00008 $0.00185
Haiku 4.5 $0.00004 $0.00092

Measured 2d ago against content hash 202f53ad5c4b, 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 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.

.claude/skills/setup-agentic-loop/templates/agentic-loop/SKILL.md · 53 lines

How it starts

The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agentic loop

The developer's attention is the scarce resource, and this loop spends it at the specification rather than during execution. They approve what is to be built; the work is then written, verified, and judged by a reviewer in a fresh context, and they hear from you once.

Reach for it when the goal is well enough understood to specify. When the developer needs to hold the change in their head as it happens, this is the wrong tool — it deliberately takes that away.

Invariant: one scheduled interruption. The front gate below is the only planned stop. After it, the loop reports when it is done — or halts and reports when it cannot proceed without guessing.

Invariant: this skill never writes code. It gates and hands off. Implementation belongs to implementer, judgement to reviewer, sequencing to orchestrator.

1. Front gate

Everything this task produces lives in one run directory: {{WORK_DIR}}/<goal-slug>/. Nothing about the task is written anywhere else, and no other task writes into it — one directory the developer can open, read in order, and delete when the task is done.

Slugify the goal and look for that directory first. It already exists with a spec → this is a resume: read the spec and progress.md, and continue from the first slice that has not landed. Do not re-derive what the record already holds.

Otherwise, produce two things in this conversation, together:

  • A spec with numbered clauses — each one a single checkable statement about behaviour after the change. Clauses are what the reviewer renders a verdict against, so a clause nobody can check is a clause nobody will enforce. If a spec-writing skill is installed, invoke it and number the criteria it produces; otherwise draft the spec here.
  • An ordered slice list — each slice independently verifiable, and each naming the clauses it satisfies.

Ask about anything you would otherwise guess at, and resolve it before the gate. Every clause of the spec is a thing three agents will act on unattended; an ambiguity that survives this gate is one they will resolve without the developer.

Read the full file on GitHub · 53 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 2d ago First seen · 53 lines · 38 tokens per session scan A 202f53ad5c4b

Subscribe to this mod's changes

agentic-loop is a skill published in the GitHub repository asteasolutions/ai-toolkit (5 stars, last pushed 6d ago), licensed MIT. It adds 38 tokens to every session and 923 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens