cancel

A command for stopping an active autoresearch loop, an automated cycle that edits an artifact, scores the result, and repeats. It removes the loop's state file while leaving the separate worktree containing the result.

In plain words
What is it for?
Checking whether a loop is active, cancelling it, and locating the isolated branch or worktree where its experiments were kept.
Why use it?
It lets another session stop a loop that is busy continuing itself, without automatically deleting work that may still need review.

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/fradser/dotclaude/cancel
Clone the repo
git clone --depth 1 https://github.com/FradSer/dotclaude
Per session 5 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 374 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 $0.00005 $0.00374
Opus 5 $0.00003 $0.00187
Sonnet 5 $0.00001 $0.00075
Haiku 4.5 $0.00001 $0.00037

Measured 2d ago against content hash 63343c8a5afb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cancel 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.

autoresearch/commands/cancel.md · 22 lines

What it actually says

Cancel Autoresearch

Run this from a separate Claude Code session in the same project directory — the looping session is busy being re-prompted by the stop hook and cannot run this itself. Removing the state file makes the loop's next stop-hook fire find no state and exit cleanly.

To cancel the autoresearch loop:

  1. Check if .claude/autoresearch.local.md exists: run test -f .claude/autoresearch.local.md && echo "EXISTS" || echo "NOT_FOUND"

  2. If NOT_FOUND: report "No active autoresearch loop found."

  3. If EXISTS:

    • Read .claude/autoresearch.local.md to get iteration: (experiment count), run_tag:, and worktree_dir: from the frontmatter
    • Remove the file: run rm .claude/autoresearch.local.md (the loop's next stop-hook fire then exits cleanly)
    • Do NOT delete the worktree automatically — it may hold an unconfirmed result. Report: "Cancelled autoresearch loop for run tag '' (was at experiment N). The result is in the isolated worktree <worktree_dir> on branch autoresearch/<tag>. To land it: cd there, review, git reset --soft <baseline> then /git:commit. To discard it: git worktree remove --force <worktree_dir> and git branch -D autoresearch/<tag>."
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 · 22 lines · 5 tokens per session scan A 63343c8a5afb

Subscribe to this mod's changes

cancel is a command published in the GitHub repository FradSer/dotclaude (588 stars, last pushed 21d ago), licensed MIT. It adds 5 tokens to every session and 374 once invoked, about $0.0000 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.