Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Verify Open Pstack

ASecurity

Verify Open Pstack through installed Codex and Claude Code skills and its shipped Bun CLIs. Use for real behavior proof, routing or setup changes, orchestration bookkeeping, and release evidence.

5 stars
0 votes
0 copies
0 views
Added 10/1/2026
ai-agentspythonshellbashnodegitdatabase

Works with

claude codecli

Security Analysis

A93/100
highPerforms destructive filesystem operations

Pro scans all 11 files and shows the line behind each finding

Scanned 10/1/2026

$npx -y skills add arjitj2/open-pstack --skill verify-open-pstack --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Verify Open Pstack?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Verify Open Pstack
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/arjitj2-verify-open-pstack/badge)](https://www.skillsdirectory.com/skills/arjitj2-verify-open-pstack)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: verify-open-pstack
description: Verify Open Pstack through installed Codex and Claude Code skills and its shipped Bun CLIs. Use for real behavior proof, routing or setup changes, orchestration bookkeeping, and release evidence.
---

# Verify Open Pstack

Read [the feature map](features/README.md) first. The primary product is the installed Pstack workflow in Codex and Claude Code. The secondary surfaces are the shipped CLI tools. A CLI proof establishes only that CLI's behavior, never installed skill discovery, parent decisions, or model execution.

This is repository maintenance tooling, outside `plugins/pstack`. Track work in GitHub Issues; this skill originated in [issue 26](https://github.com/arjitj2/open-pstack/issues/26). Read `UPSTREAM.md` before modifying upstream-derived content.

## Launch

Run from the Open Pstack checkout in one Bash session. There is no server, port, database seed, or build step. Bun installs the CLI dependencies; Python 3 checks observed JSON. Installation may access the package registry. Each CLI invocation exits on its own.

```bash
set -euo pipefail
VERIFY_REPO=$(git rev-parse --show-toplevel)
VERIFY_TOOLS="$VERIFY_REPO/plugins/pstack/skills/poteto-mode/scripts"
VERIFY_EVIDENCE=$(mktemp -d "$HOME/open-pstack-verification.XXXXXX")
VERIFY_SCRATCH=$(mktemp -d "${TMPDIR:-/tmp}/open-pstack-verify.XXXXXX")
export VERIFY_REPO VERIFY_TOOLS VERIFY_EVIDENCE VERIFY_SCRATCH
cleanup() {
  if [ -d "$VERIFY_SCRATCH" ]; then
    cp -R "$VERIFY_SCRATCH" "$VERIFY_EVIDENCE/scratch-final"
    rm -rf -- "$VERIFY_SCRATCH"
  fi
}
trap cleanup EXIT
capture() {
  local label=$1 rc=0
  shift
  printf '%q ' "$@" > "$VERIFY_EVIDENCE/$label.command"
  printf '\n' >> "$VERIFY_EVIDENCE/$label.command"
  "$@" > "$VERIFY_EVIDENCE/$label.stdout" 2> "$VERIFY_EVIDENCE/$label.stderr" || rc=$?
  printf '%s\n' "$rc" > "$VERIFY_EVIDENCE/$label.exit"
  cat "$VERIFY_EVIDENCE/$label.stdout"
  cat "$VERIFY_EVIDENCE/$label.stderr" >&2
  return "$rc"
}
printf '%s\n' "$VERIFY_EVIDENCE"
capture revision git rev-parse HEAD
capture initial-status git status --short
capture plugin-before git ls-files -s plugins/pstack
capture install bun install --frozen-lockfile --cwd "$VERIFY_TOOLS"
```

Readiness means install exits zero and Doctor passes. For parent-app checks, use the installation feature's separate launch instructions. Real providers require their existing authenticated subscriptions and the saved model sheet's spending policy. Do not print credentials, invent access, or run provider probes just to verify a local CLI.

## Doctor

This single read-only command checks the checkout identity, matching manifest versions, installed CLI dependency, runtime, and actual CLI entry point. Run it after Launch and again if anything looks wrong. It checks CLI readiness, not parent authentication.

```bash
capture doctor bun -e '
const root = process.env.VERIFY_REPO;
const tools = process.env.VERIFY_TOOLS;
const a = await Bun.file(`${root}/plugins/pstack/.claude-plugin/plugin.json`).json();
const b = await Bun.file(`${root}/plugins/pstack/.codex-plugin/plugin.json`).json();
if (a.name !== "pstack" || b.name !== "pstack" || a.version !== b.version) throw Error("manifest mismatch");
if (!await Bun.file(`${tools}/node_modules/commander/package.json`).exists()) throw Error("run Launch first");
const p = Bun.spawnSync([process.execPath, `${tools}/orch/orch.ts`, "--help"]);
if (p.exitCode !== 0 || !p.stdout.toString().includes("Plain-file orchestrate bookkeeping")) throw Error("CLI unavailable");
console.log(JSON.stringify({root, version:a.version, bun:Bun.version, surface:"checkout CLI"}));'
```

## Drive

Choose the changed feature, then execute its recipe with the variables and `capture` function above. Start with [orchestration](features/orchestration.md) for a self-contained live CLI smoke. Use fresh scratch and evidence directories for every attempt; never use the user's default orchestrate store. Separate CLI stores can run concurrently, but dependency installation should finish before any CLI drive begins.

Parent installations, authentication, and model sheets are shared state unless a dedicated test profile is provided. Do not double-drive a parent session or change a personal model sheet for a fixture. When no isolated authenticated parent is available, record that entry point as blocked; continue independent CLI coverage.

## Evidence

`$VERIFY_EVIDENCE` is a persistent directory under the user's home, separate from disposable scratch. Every command produces `.command`, `.stdout`, `.stderr`, and `.exit` files. Use unique capture labels. Preserve fixture bytes, before/after model sheets without secrets, runner receipts, native host events, and parent transcripts or screenshots as applicable.

Record feature ID, entry point, checkout commit and dirty diff, installed version and candidate-tree identity for parent runs, action, expected result, observed result, and skips in `result.md` in that directory. Capture the action and resulting state. Confirm writes through a second CLI read and stored files; final agent prose alone is not proof. Never fabricate provider failures or count synthetic policy inputs as a real fallback execution. Mocks belong only at an existing production boundary and must be labeled. Do not assume a dry-run is offline or side-effect-free: observe files, git refs, and network activity for the behavior being claimed.

Before release, install the exact candidate and exercise changed behavior in **every affected parent**. Unit tests, source checks, CLI policy decisions, and model self-reports cannot replace this gate. Follow `.github/pull_request_template.md`; missing applicable live evidence means draft. For maintenance-only changes, prove `plugins/pstack` unchanged and exercise the actual changed maintenance path instead.

## Cleanup

Run even after a failed drive. The EXIT trap covers failures in the Bash session; explicitly run the following on success. The copied scratch state survives for inspection. Keep the evidence directory.

```bash
capture plugin-after git ls-files -s plugins/pstack
capture plugin-index-equality diff -u "$VERIFY_EVIDENCE/plugin-before.stdout" "$VERIFY_EVIDENCE/plugin-after.stdout"
capture plugin-working-tree git diff --exit-code HEAD -- plugins/pstack
capture plugin-untracked git ls-files --others --exclude-standard -- plugins/pstack
test ! -s "$VERIFY_EVIDENCE/plugin-untracked.stdout"
cleanup
test ! -e "$VERIFY_SCRATCH"
test -s "$VERIFY_EVIDENCE/doctor.stdout"
printf 'Scratch removed; evidence retained at %s\n' "$VERIFY_EVIDENCE"
trap - EXIT
```

Tree-equality checks above apply to maintenance-only work and should fail if packaged files changed. For intentional plugin changes, retain the diff and use installed-candidate proof instead. Close only parent sessions created for this run and remove only their dedicated test profiles. Never kill by process name or delete shared plugin caches. Foreground CLI checks here leave no background process.

## Helpers

No helper executable is shipped. Launch defines the shell functions used throughout. The product executables are `bun "$VERIFY_TOOLS/orch/orch.ts"`, `"$VERIFY_TOOLS/model-policy/pstack-model-policy"`, and `"$VERIFY_TOOLS/runner/pstack-runner"`; use the feature recipes for arguments.

Keep this map current with `pstack:maintain-verification-skill` (`/pstack:maintain-verification-skill` in Claude Code).

Attribution

arjitj2arjitj2
View sourceSee grades on GitHubMore from arjitj2 →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698431 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →