AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Home/
Dev Tooling/
pstack-claude
pstack-claude logo

pstack-claude

Active·★ 4.4k·MIT·Updated 2026-11-17

michael-denyer/pstack-claude is an open-source (MIT licensed) high-fidelity port of Lauren Tan's opinionated 'pstack' (poteto-stack) Cursor agent skills, custom-engineered for Claude Code, Codex, Pi, GitHub Copilot CLI, and Muse. It provides a disciplined engineering harness that reins in LLM verbosity and enforces strict code quality.

Unlike upstream installations that force agents to translate Cursor-specific syntax at runtime—wasting prompt tokens and introducing hallucination risk—this port translates skills ahead of time and injects native SessionStart hooks. Activating 'poteto-mode' guides the agent toward minimal, self-contained implementations without redundant defensive checks. Furthermore, it incorporates formal verification tooling (/create-formal-verification-skill) to tackle critical concurrency invariants using TLA+, Lean proofs, and Kani for Rust.

pstack-claude installs seamlessly through Claude Code's plugin marketplace or Codex CLI. Developers adopting it should embrace its unapologetically minimalist philosophy: it actively discourages boilerplate abstractions and defensive over-engineering, which may require cultural adjustment for teams accustomed to traditional object-oriented conventions.

This write-up is based on the project's README and public repository data. We have not hand-tested the tool. Data as of 2026-11-12.

License
MIT
Latest release
v0.9.136 · 2026-11-11
Last commit
2026-11-11
Contributors
34
Forks
390
Open issues and PRs
6
First published
2026-05-26

Source: GitHub, npm and PyPI public APIs.

#agent-plugin#agent-skills#agentic-ai#anthropic#chatgpt#claude#claude-code#claude-code-skills
↗ Visit site★ GitHub
01

Features

01Routes tasks through poteto-mode using hooks or host-specific extensions.
02Provides playbooks for bugs, features, refactoring, performance and shipping.
03Offers setup controls for routing, default models and reasoning effort by role.
04Includes a workflow for creating a reusable formal-verification project skill.
02

Why choose it

+Pre-compiled native translation for Claude Code eliminates token waste and runtime translation bugs found in upstream.
+Enforces 'poteto-mode' to produce remarkably concise, minimal, and verifiable code without boilerplate bloat.
+Integrates cutting-edge formal verification skills for TLA+ model checking, Lean proofs, and Kani Rust verification.
+Wide multi-agent host support across Claude Code, Codex, Pi, GitHub Copilot CLI, and Muse.
03

Trade-offs

!Strict minimalism and rejection of defensive boilerplate may conflict with legacy corporate coding standards.
!Formal verification workflows require specialized external tools (TLA+ tools, Lean 4, Kani) installed locally.
!Relies on Claude Code plugin architecture and third-party marketplace updates to stay in sync with upstream changes.
!Requires developer familiarity with opinionated slash commands (/create-formal-verification-skill, poteto-mode).
04

Maintenance check

ActiveChecked: 2026-11-12
Last commit
2026-11-11
Commits, last 90 days
100+
Releases, last 90 days
30
Latest release
v0.9.136 (2026-11-11)
Open issues
6
Recent issues with at least one comment
4/10
Commits per week, last 12 weeks
Commits per week, last 12 weeks: 20, 7, 13, 9, 16, 59, 72, 20, 13, 21, 66, 20

Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.

Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.

Most discussed open issues

  • #310 Opt-in cross-provider model reviewers for interrogate (e.g. Claude Opus 5.5 vs. GPT-6.1 Sol)2 comments · 2026-10-23
  • #344 MiniMax Code as a harness: a pstack port, and the three places this tree needs a MiniMax row1 comments · 2026-10-31
  • #450 Eval playbook: ship a runner for before/after skill trials0 comments · 2026-11-11
  • #347 Seven correctness fixes carried locally against 0.9.114: interest in PRs?0 comments · 2026-11-01

Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.

05

Use cases

↳Reproduce a bug, investigate its cause and verify the resulting fix.
↳Apply a consistent skill workflow across supported coding-agent environments.
↳Capture a checked invariant in a project skill for later verification.
06

Alternatives

⚡Top VS:
VScoder_eval★160→VSClaude Code Templates★33.9k→VSChisle★1.2k→VSdarwin-skill★7.0k→
coder_eval logo
coder_eval★ 160
Playwright for coding agents. Test that your skills, MCP servers, and CLIs actually work when an agent uses them — sandboxed YAML suites, A/B experiments, CI gates.
VSCompare→
Claude Code Templates logo
Claude Code Templates★ 33.9k
Open-source registry and CLI for Claude Code components: agents, commands, skills, hooks, MCPs, settings and mods
VSCompare→
Chisle logo
Chisle★ 1.2k
Cut your AI coding agent's token bill on three axes: terse prose, YAGNI-first code, and tool-output compression. Claude Code, Pi, Cursor, Codex, Gemini + 4 more. Zero deps, published benchmarks including the runs it loses.
VSCompare→
darwin-skill logo
darwin-skill★ 7.0k
达尔文.skill —— 一个让你的Skill无限进化的系统:评估→改进→测试→保留或回滚 | Autoresearch-inspired autonomous skill optimization for Claude Code. Evaluate, improve, test, keep or revert.
VSCompare→
hermes-agent logo
hermes-agent★ 264.2k
The agent that grows with you
VSCompare→
skills logo
skills★ 4.0k
Agent Skills for NVIDIA products — install into Claude Code, Codex, and other coding agents to run Physical AI, robotics, simulation, CUDA, and RAG workflows end to end.
VSCompare→
SkillEvaluator logo
SkillEvaluator★ 3.3k
Multi-tier framework for evaluating AI agent skills with quality gates, semantic overlap detection, synthetic evaluation dataset generation, and live agent evaluation that measures how skills affect agent behavior.
VSCompare→
delegate-skills logo
delegate-skills★ 2.6k
Delegate a coding task to a separate coding agent CLI, review the diff, land the commit yourself — one per implementer.
VSCompare→
See all alternatives →

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Why choose it03Trade-offs04Maintenance check05Use cases06Alternatives
michael-denyer /
pstack-claude
↗

Opinionated agent skill stack for Claude Code, Codex, and Copilot emphasizing concise, verified code and formal verification

4.4k390MIT
Last commit1d ago
StatusActive
CategoryDev Tooling
Contributors34
Latest releasev0.9.136
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.