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

Factory

ASecurity

Use to turn an intention into a bounded, verifiable factory order with a machine-checked receipt.

4 stars
0 votes
0 copies
1 views
Added 9/19/2026
ai-agentsgo

Works with

cli

Security Analysis

A100/100

Scanned 9/19/2026

$npx -y skills add mverab/slashstack-skills --skill factory --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Factory?

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

Security grade badge for Factory
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/mverab-factory/badge)](https://www.skillsdirectory.com/skills/mverab-factory)

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: factory
description: Use to turn an intention into a bounded, verifiable factory order with a machine-checked receipt.
---

# Factory

## Purpose

Turn an intention into a bounded, verifiable factory order. The order freezes
what "correct" means before the work starts; the receipt proves what actually
ran, against which repository state, with which result. A green test alone is
never "done" — the acceptance criteria bound to the order decide.

## When to Use

Use when a change should be delivered as a verifiable unit: an export filter,
a bug fix with a regression test, a refactor with an observable contract.
Do not use it as a task queue or an autonomous scheduler — this vertical
delivers one bounded, verified change at a time.

## Workflow

1. Initialize once per repo: run `npx slashstack factory init`. It detects the
   test command and creates `.agents/factory/`. It never edits application code.
2. Define the goal first: if the work is multi-step, register it in
   `.agents/memory/goal-registry.json` with the execute workflow. The factory
   order is the delivery contract for that goal, not a replacement for it.
3. Create the order: `npx slashstack factory order create --title "<what>" --acceptance "<observable criterion>"`.
   Add `--bind-test` only when the frozen test command IS the acceptance check.
   Without it, verification stays blocked with `acceptance_not_evaluated` —
   that is the system refusing to call untested work "done".
4. Implement the change inside the order's intent. Keep the diff scoped.
5. Verify: `npx slashstack factory verify <order-id>`. It runs the frozen
   verifier and writes a receipt under `.agents/factory/receipts/` with the
   exact commit, duration, exit code, and redacted output.
6. Or let the configured agent implement it: set `agent.command` in
   `.agents/factory/config.json` (any agent CLI; it receives the order via
   FACTORY_ORDER_* env vars), then `npx slashstack factory run <order-id>`.
   The agent works in an isolated worktree under `.agents/factory/runs/` and
   the result is verified automatically. Your main checkout is never touched.
7. Read the receipt honestly: `failed` means repair; `blocked` means the
   contract itself changed (verifier, worktree mid-check, missing test command)
   or the run could not start (no agent command, another run active) —
   that needs a human decision, not a retry loop.
8. Ship through the normal `ship` workflow. Merge and deploy stay human
   decisions.

## Output

A receipt JSON with `status` (`passed` / `failed` / `blocked`), the
pre-check repository identity, the frozen command, and redacted evidence.
A `passed` receipt means: the checks named by the order ran green on the
recorded repository state. It does not prove the product is good — only that
the layer a machine can check is intact.

Attribution

mverabmverab
View sourceSee grades on GitHubMore from mverab →
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', ...

698621 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 →