pr-cockpit: Instructions file for Codex

AGENTS.md

pr-cockpit AGENTS.md is an instructions file for Codex, OpenCode from theolundqvist/pr-cockpit. It costs 4,128 tokens per session, scanned A, original, MIT.

A set of contributor instructions for PR Cockpit, an app for reviewing and managing GitHub pull requests. It describes the server, Svelte user interface, Electron desktop shell, webhook relay, and command-line scripts, along with local development and layout rules.

In plain words
What is it for?
Use it when changing GitHub data handling, the SQLite cache, review actions, the web interface, desktop shell, webhook relay, CLI, or screenshot tests. It explains how to start the interface and run targeted tests.
Why use it?
It gives agents the project structure and exact commands needed to run or test the relevant part of the app. It also records interface dimensions and layout constraints that help prevent visual regressions.

Instructions file for CodexOpenCode

Written for Codex and OpenCode: the file is AGENTS.md.

This is theolundqvist/pr-cockpit's own configuration. It tells Codex and OpenCode how to work on pr-cockpit itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything pr-cockpit configures →

Not installable: its command points at a path on the author’s own machine, so it runs nowhere else. The line is /Users/theo/dev/pr-cockpit.

Reuse

Borrowing it

Nothing to install: this file belongs to theolundqvist/pr-cockpit. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/theolundqvist/pr-cockpit/main/AGENTS.md
Clone the repo
git clone --depth 1 https://github.com/theolundqvist/pr-cockpit

Made for: Codex, OpenCode.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for pr-cockpit AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/theolundqvist/pr-cockpit/agents-md/github.svg)](https://agentmods.dev/instructions/theolundqvist/pr-cockpit/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/theolundqvist/pr-cockpit/agents-md"><img src="https://agentmods.dev/badge/instructions/theolundqvist/pr-cockpit/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for pr-cockpit AGENTS.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/theolundqvist/pr-cockpit/agents-md"><img src="https://agentmods.dev/badge/instructions/theolundqvist/pr-cockpit/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 4,128 This file is loaded in full into every session.
When invoked 4,128 The same file — it is already loaded in full.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.04128 $0.04128
Opus 5.5 $0.01651 $0.01651
Sonnet 5.5 $0.00826 $0.00826
Haiku 4.5 $0.00413 $0.00413

Measured yesterday against content hash 11f210e79f69, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

pr-cockpit AGENTS.md scanned grade A with 1 finding 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

3. Verify with `curl -fsS -i http://127.0.0.1:<port>/healthz`. A healthy response is `200` with `root`, `lastPollAt`, and `prCount`. Note that `lastPollAt` is `null` immediately after a restart until the first poll compl
AGENTS.md · 129 lines

How it starts

The opening of the file, as written. The whole thing — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.

PR Cockpit contributor and agent guide

This repository is worked on by humans and coding agents alike. Leave it easier to operate than you found it.

Layout

  • server/: the Bun server. GitHub ingestion, SQLite cache, mutation queue, HTTP API.
  • ui/: the Svelte 5 front end (runes: $state, $derived, $props, $effect).
  • shell/: the Electron shell. It only provides the macOS window, global hotkeys, and prcockpit:// links.
  • relay/: the Cloudflare Worker that receives GitHub webhook events and hands them to a cockpit.
  • scripts/: install, launch, update, and the pr-cockpit CLI.

Working here

  • Always work directly on main and pull origin/main before starting work.
  • Commit continuously in small, coherent increments and push every commit immediately to origin/main.
  • Never create pull requests or feature branches for work in this project. This project-specific workflow overrides generic instructions to open a pull request.
  • bun install at the repository root, in ui/, and in shell/.
  • cd ui && bun run dev for a hot-reloading UI. It proxies /api to 127.0.0.1:4820, so it needs a server on that port. Run scratch previews with the installed UI Vite under Node; root bunx downloads unlocked dependencies, and Bun lacks the proxy’s socket.destroySoon.
  • bun test server/<file>.test.ts for a targeted test. Prefer targeted runs over the whole suite while iterating.
  • Screenshot harnesses render at 1600x1200; keep their default viewport and PNG dimension checks aligned.
  • PR detail headers and tabs use the standard width; Conversation centers its 816px primary column and adjacent sidebar beneath them, and Files alone expands to full width.
  • Agents is a conversation workspace, not a status dashboard: prioritize session discovery and transcript reading; keep history and logs secondary, because repeated badges and competing controls obscure the work.
  • Whiteboard scroll zooms at the cursor; Space-drag pans. Seed layouts fit the canvas proportions. Only Arrange refreshes inbox-generated group membership, reflows, and removes empty sections, undoably; custom sections stay personal. Otherwise preserve positions. Show tool keys beside their labels.
  • Verify landing-page search queries against live GitHub results; captured PR titles can disappear from the search index while the fixture remains valid.
  • Match existing style. Keep the homepage dark and speed-first, with brief subtext and FAQ answers; link to details. Comments explain hidden constraints.
  • Pull-request actions follow GitHub semantics: merge green, update neutral, and destructive actions red.
  • Pins are bookmarks, not merge consent. Approved for safe merge is explicit PR-wide permission exposed as approvedForSafeMerge in CLI JSON; re-read it before merging, never grant it yourself, and retain all safety checks. Approval survives pushes but ends on revoke, close, or feature disable; replicas use the source's approval state.
  • Cross-group inbox drops change the grouping's source: conventional-commit type or scope via the title mutation queue, or personal manual assignment. Keep draft markers, breaking markers, and the rest of the title intact. Computed status sections are never writable by dragging.
  • Workflow jobs use commit-check semantics: green check for success, red cross for failure, orange spinner while active, and gray minus when skipped. Logs preserve ANSI colors and open at the bottom. Commit status icons and failing-check rows open Cockpit’s Actions tab at the exact commit and job; its selector covers the full mirror range, not GraphQL’s last 100 commits.
  • Operational telemetry belongs on a dedicated Usage page in Settings, not among general controls; it keeps three days of hourly context and forecasts the active GitHub quota window.
  • File status lives in the colored leading icon; moved files use a distinct move glyph, never a text badge beside the name. 1: - Agents mutate existing PRs through pr-cockpit; use its --body-file commands for exact multiline text, never gh or direct GitHub APIs.
  • Agent reads share a per-PR 60-second GitHub revalidation window, including fresh/listen reads and --new-since queries. A newer webhook permits one immediate revalidation; concurrent reads share in-flight work. Cached responses retain their actual snapshot timestamp and freshness. With REST recording enabled, agent JSON's quota.hotPrs ranks personal core calls over the last 60 minutes by PR and dominant source; quota.identities[].hotPrs separates installation usage. Markdown and exhausted-quota errors identify the hottest PR for the affected credential.
  • Keep examples, fixtures, benchmarks, and docs free of private company, product, repository, host, and infrastructure names; do not perform unrelated hosting or infrastructure work here. 2: COCKPIT_PROXY="build-server" # optional Cockpit replica source over SSH 3: # COCKPIT_PROXY="build-server" # replicate another Cockpit over SSH
  • For manual app recordings, launch /Users/theo/dev/pr-cockpit; this pr-cockpit-v3 checkout is the legacy landing worktree.

Read the full file on GitHub · 129 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 Changed · +1 lines · +85 tokens per session 11f210e79f69
  2. 4d ago Changed · +1 lines · +5 tokens per session d118676cbf8b
  3. 7d ago Changed · +2 lines · +123 tokens per session 69115a2d72b4
  4. 8d ago Changed · +14 tokens per session e9a6421fe43e
  5. 23d ago Changed · +49 tokens per session 52ab6f2e433d
  6. 24d ago Changed · +1 lines · +150 tokens per session f8d456411974
  7. 27d ago Changed · +1 lines · +134 tokens per session 653ac66e91f1
  8. 28d ago Changed · +1 lines · +65 tokens per session c48faff1ee27
  9. 29d ago Changed · +2 lines · +162 tokens per session 55f0f368d07e
  10. 1mo ago First seen · 120 lines · 3,341 tokens per session scan A 4efd057491c5

Subscribe to this mod's changes

pr-cockpit AGENTS.md is an instructions file published in the GitHub repository theolundqvist/pr-cockpit (51 stars, last pushed today), licensed MIT. It adds 4,128 tokens to every session, about $0.0165 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other instructions, from other repositories

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 9,715 tokens

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,461 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).

microsoft/vscode · 5,262 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,680 tokens

vscode copilot-instructions.md

Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).

microsoft/vscode · 2,750 tokens

deno copilot-instructions.md

Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.

denoland/deno · 3,077 tokens