stop

A command for stopping an autoresearch run, an automated loop that repeatedly tests changes. It records that the run stopped and writes a final report without deleting the experiment's work.

In plain words
What is it for?
Use it when you want to halt autoresearch, stop the loop, or finish an interrupted run with its status and recovery information recorded.
Why use it?
It provides a clear stopping point and preserves the run's state so the work can be reviewed or resumed later.

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/emaballarin/ccplugins/stop
Any agent
npx skills add emaballarin/ccplugins --skill stop
Clone the repo
git clone --depth 1 https://github.com/emaballarin/ccplugins

Made for: Claude Code, Codex.

Per session 106 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 608 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 $0.00106 $0.00608
Opus 5 $0.00053 $0.00304
Sonnet 5 $0.00021 $0.00122
Haiku 4.5 $0.00011 $0.00061

Measured yesterday against content hash 12d2438d5f50, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

stop 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.

plugins/autoresearch/skills/stop/SKILL.md · 51 lines

What it actually says

/ar:stop — end the run

Halt cleanly, leaving everything recoverable.

Steps

  1. Read state (cat ./.ar/ar.jsonl | tail -50). No ar.jsonl means no run to stop — say so and stop.

  2. If HEAD carries a Result: pending trailer, an iteration was interrupted mid-measurement. Leave the commit alone, and record it in the sentinel's reason so a later /ar:resume re-measures rather than reverting it.

  3. Append the sentinel — the last line of the file, and what ar-loop.sh greps for:

    { "run": <n>, "status": "stopped", "segment": <s>, "reason": "<why>" }
    
  4. Write ./.ar/final_report.md (see the /ar:report skill for its shape).

  5. Print the final status block, the best commit, and — as text, not as a command to run — how to get back:

    git switch <config.originBranch>          # return to where the run started
    git log --format='%h %s' --grep='"status":"keep"' <config.branch>
    

Hard rules

  1. Destroy nothing. No reverts, no branch deletion, no git clean. Stopping is not discarding: the experiment branch and every kept commit stay exactly as they are, and the operator decides later what to merge or bin.
  2. Do not switch branches. Print the command and leave the experiment branch checked out for inspection.
  3. Stopping is reversible — a later /ar:resume opens a new segment and carries on from the same state. Say so.

Completion status

End with a terminal status token as the last line of your reply — DONE, DONE_WITH_CONCERNS, BLOCKED, or NEEDS_CONTEXT — per references/completion-status.md. For /ar:stop, DONE once the status:stopped sentinel is appended and the report written; NEEDS_CONTEXT if there is no active run to stop.

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 · 51 lines · 106 tokens per session scan A 12d2438d5f50

Subscribe to this mod's changes

stop is a skill published in the GitHub repository emaballarin/ccplugins (3 stars, last pushed 26d ago), licensed MIT. It adds 106 tokens to every session and 608 once invoked, about $0.0005 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

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 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

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens