stopping-runaway-prs
Use when a pull request has grown well past its estimate, review rounds keep finding new defects, fixes keep spawning more fixes, or a change that claims to…
Skill by basnijholt
Chalk Labs
Need help deploying stopping-runaway-prs?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/basnijholt-stopping-runaway-prs
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/stopping-runaway-prs
What's inside
- Skill name
stopping-runaway-prs- Description
- Use when a pull request has grown well past its estimate, review rounds keep finding new defects, fixes keep spawning more fixes, or a change that claims to simplify code adds code
- License
- MIT
- Instructions
- 20 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Stopping Runaway PRs Growth and recurring review findings are design signals. When net production LOC exceeds twice the estimate, a third review round finds new defects, or a "simplification" adds code, audit before writing any fix: 1. Measure net production LOC against the estimate (`git diff --numstat <base>...HEAD`, production files only). 2. Cluster findings from all rounds; a shared root means the boundary is wrong. 3. Give options with net LOC each: keep patching, restructure around the root, shrink to the original goal, or restart clean. 4. Send your human partner the numbers and one recommendation in under 150 words, and wait for the choice before patching. "Fix these and merge" was said before they saw the numbers. ## Red Flags | Thought | Reality | |---|---| | "Doing the four fixes now, and I'll flag the risk alongside" | Patching first is the next round of the same loop. Audit first. | | "Ship it behind a flag for the deadline, redesign after" | That is one option to price in the message, not a decision to make by patching. | | "If the next round finds more, then we stop" | Every earlier round could have said that. Stop now. |
From basnijholt/baspowers, MIT.
More from basnijholt/baspowers
- 0.592score
executing-plans
basnijholt
Use when executing an implementation plan in the current session as the implementer yourself — Native (inline) execution was selected for the plan, or no…
SkillAFreshness: A★ 1Claude Code +9 - 0.592score
diagnosing-baspowers
basnijholt
Use when a baspowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't…
SkillAFreshness: A★ 1Claude Code +9 - 0.591score
using-baspowers
basnijholt
Use when starting a new conversation or task, or when the task changes enough that different skills may apply
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
merging-when-approved
basnijholt
Use when asked to review a pull request and merge it only if approved, or before merging a pull request you wrote or changed
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
consulting-cross-model
basnijholt
Use when a hard technical decision remains ambiguous after inspecting relevant evidence and choosing incorrectly would cause meaningful rework
SkillAFreshness: A★ 1Claude Code +9 - 0.586score
triaging-pr-backlogs
basnijholt
Use when asked to clean up, triage, or prune a backlog of open pull requests or issues
SkillAFreshness: A★ 1Claude Code +9 - 0.586score
scoring-pull-requests
basnijholt
Use when asked whether a pull request or change is good, clean, big, risky, over-engineered, or worth merging
SkillAFreshness: A★ 1Claude Code +9 - 0.582score
reporting-status
basnijholt
Use when asked where things stand, what is left, or whether work is done or stuck, and when ending a turn after work on several items
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
wait-what
basnijholt
Use when the last message did not land and needs a simpler re-pitch with the missing context
SkillAFreshness: A★ 1Claude Code +9 - 0.564score
brainstorming
basnijholt
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior.
SkillAFreshness: A★ 1Claude Code +9 - 0.562score
receiving-code-review
basnijholt
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires…
SkillAFreshness: A★ 1Claude Code +9 - 0.562score
using-git-worktrees
basnijholt
Use when starting feature work that needs isolation from the current workspace, or before executing implementation plans
SkillAFreshness: A★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 8, 2026 (23 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/basnijholt/baspowers
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from basnijholt
- 0.592score
executing-plans
basnijholt
Use when executing an implementation plan in the current session as the implementer yourself — Native (inline) execution was selected for the plan, or no…
SkillAFreshness: A★ 1Claude Code +9 - 0.592score
diagnosing-baspowers
basnijholt
Use when a baspowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't…
SkillAFreshness: A★ 1Claude Code +9 - 0.591score
using-baspowers
basnijholt
Use when starting a new conversation or task, or when the task changes enough that different skills may apply
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
merging-when-approved
basnijholt
Use when asked to review a pull request and merge it only if approved, or before merging a pull request you wrote or changed
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
consulting-cross-model
basnijholt
Use when a hard technical decision remains ambiguous after inspecting relevant evidence and choosing incorrectly would cause meaningful rework
SkillAFreshness: A★ 1Claude Code +9 - 0.586score
scoring-pull-requests
basnijholt
Use when asked whether a pull request or change is good, clean, big, risky, over-engineered, or worth merging
SkillAFreshness: A★ 1Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get stopping-runaway-prs running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
stopping-runaway-prs skill. Agent Mart. https://agentmart.dev/skills/basnijholt-stopping-runaway-prs (last updated 2026-10-24)