pr-main

A release-preparation procedure for changing a development version into a stable software release and opening its release pull request.

In plain words
What is it for?
Use it to bump the version, update release notes, run release checks, and open one development-to-main pull request.
Why use it?
It checks that development is ready, updates version and changelog records, and stops before the final merge and publication decisions.

Command for Claude Code

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/littlebearapps/untether/pr-main
Clone the repo
git clone --depth 1 https://github.com/littlebearapps/untether

Made for: Claude Code.

Per session 99 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,807 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 $0.00099 $0.01807
Opus 5 $0.00049 $0.00903
Sonnet 5 $0.00020 $0.00361
Haiku 4.5 $0.00010 $0.00181

Measured 2d ago against content hash d5381771150b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

pr-main 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/commands/pr-main.md · 136 lines

How it starts

The opening of the file, as written. The whole thing — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are handling /pr-main. /pr-main prepares a stable release and opens the devmaster PR — then STOPS. Merging that PR is Nathan's single release gate (→ auto-tag-on-master.ymlrelease.yml → PyPI). /pr-main does everything Claude is allowed to do and hard-stops at the operator boundary.

User input: $ARGUMENTS

Untether adaptations (read first)

Load .claude/rules/workflow-commands.md (routing + cross-cutting rules) and .claude/rules/release-discipline.md (semver, changelog format, FAQ, staging vs release). Key points:

  • Authority boundary (verified against release-guard.sh). gh pr create --base master is allowed (creating a PR is not a push/merge/tag/release). gh pr merge to master, git push master, git tag v*, and gh release create are blocked. So /pr-main may OPEN the release PR but may never merge/tag/release it — and it must not try (don't fight the guard).
  • The version-bump commit lands on dev (pushing to dev is allowed; master is not). The PR is devmaster.
  • fleet-rollout.sh is the operator's step/pr-main never runs it. The attestation marker is surfaced as advisory context in the PR body only.
  • Confirm-gated + idempotent. Surface the PR title + body and wait for a tap; a re-invoked /pr-main must not double-open the release PR (check for an existing open devmaster PR first).
  • Untether-mode. State the confirmation in text and STOP for a reply. Keep the report brief.

Sub-commands

Form Action
/pr-main X.Y.Z Prepare the stable release + open the devmaster PR, then STOP
/pr-main X.Y.Z --dry-run Prepare + validate + print the would-be PR body; open nothing
/pr-main --rc-summary Collapse the rc CHANGELOG sections + validate only; no bump, no PR
/pr-main --help Usage, then stop

Flow

M-1. Confirm the release is cuttable

  • git fetch and confirm dev is green + ahead of master (CI on dev is the last TestPyPI publish).
  • Confirm the intended stable X.Y.Z is decided and is a proper stable version (no rc/a/b/dev suffix — those are skipped by auto-tag-on-master.yml).
  • If an open devmaster PR already exists for this version → STOP (idempotent; don't double-open).

Read the full file on GitHub · 136 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. 2d ago First seen · 136 lines · 99 tokens per session scan A d5381771150b

Subscribe to this mod's changes

pr-main is a command published in the GitHub repository littlebearapps/untether (66 stars, last pushed 3d ago), licensed MIT. It adds 99 tokens to every session and 1,807 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 2026-08-30.