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

Dim Factory

ASecurity

Operate factory orders as the operator — run each order, judge every artifact against the record, carry out the owner's decisions and settle questions from the record or an experiment before asking the owner. Use on an order id the owner names.

5 stars
0 votes
0 copies
0 views
Added 10/5/2026
researchgo

Security Analysis

A100/100

Scanned 10/5/2026

$npx -y skills add cniska/dim-factory --skill dim-factory --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Dim Factory?

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

Security grade badge for Dim Factory
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/cniska-dim-factory/badge)](https://www.skillsdirectory.com/skills/cniska-dim-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: dim-factory
description: Operate factory orders as the operator — run each order, judge every artifact against the record, carry out the owner's decisions and settle questions from the record or an experiment before asking the owner. Use on an order id the owner names.
argument-hint: "<order>"
---

# Factory

The operator runs the line and never does a station's work. The first operator action from this session registers it as the project's operator, and every later `dim` command from it acts as that operator.

## Commands

- `dim order run <order>` runs whatever the record says comes next, the station the order is at or its ship, and returns when it finishes.
- `dim order show <order>` prints the order: its status, station, `next`, `admits`, branch, workspace, log, workers, slices and findings.
- `dim order approve <order> --reason <reason> --decided owner|operator` and `dim order return <order> --reason <reason> --decided owner|operator` carry out the decision on the artifact that waits. Approval runs the next station or ships; a return reruns the same station with the reason in its brief.
- `dim order update <order> [--title <title>] [--description <description>]` changes the order until its plan is approved, and the next run plans it again.
- `dim order cancel <order> --reason <reason>` ends the order, stopping any station working on it.
- `dim message send <text> --order <order> --to plan|build|review` runs a turn of that station's worker and prints its reply.
- `dim trace <order>` follows one order's factory steps while it runs. `dim session show <session>` prints a worker's transcript. `dim query search`, `dim query prior-art` and `dim query thread` read the record.

Intake is `dim-add`.

## Running an order

1. `dim order show <order>`. `next` is `run`, `approve` or `update`, and `admits` lists what the order takes now; any other action is refused naming them. An order whose run is alive refuses every action but cancelling a station turn.
2. `next` is `run`: `dim order run <order>`. It returns once the station has returned, or with the refusal that stopped it, and `dim order show` holds the result.
3. `next` is `approve`: an artifact waits. Judge it as below, then approve or return it.
4. `next` is `update`: the planner returned the order, and its `order_returned` entry holds why. Update the order and run it, or cancel it with the reason.
5. A failed run leaves the order where the record puts it. The refusal names its code, its cause and the command that resolves it, mostly `dim order run <order>` once a cause outside the order is cleared: a dirty checkout of the default branch, a session that died at its usage limit with `resetsAt` in the log, a role with no model in the user's config.

A `message_sent` entry from a worker in the log is a note for the operator; read it before judging.

## Judging an artifact

The artifact is the worker's explanation; the record is the evidence. Read the artifact, then check each claim against `dim order show`: the plan's slices against the description and the commits already on the branch; the Build artifact against the `slice_committed` entries and their check evidence, the branch's commits and the answers to findings; the Review artifact's `covered` against the areas `dim-review` runs and its `unverified` against what the order needs. The result names what the record confirmed and what it did not, never the worker's summary repeated.

Each artifact is held to what its station's skill asks of it, `dim-plan`, `dim-build` or `dim-review`, and a refused finding to a reason the code bears out.

Approving the Review artifact ships the order. `--decided owner` carries the owner's decision; `--decided operator` only where the owner has handed that decision to the operator. The reason is recorded with it.

## Questions

Classify a question before asking the owner. A fact the record holds is read with `dim order show`, `dim session show` or `dim query search`. A fact an experiment can show, such as how a command behaves or what an output looks like, is run in a scratch directory, never in an order's workspace, and the result decides. Only a product or preference call that no record or experiment settles goes to the owner, with the options and a recommendation.

## Result

Report the order id, its status and station, `next`, the decision taken with its reason, and what the record confirmed about the artifact. For a stop, the refusal's code and the command that resolves it.

Attribution

cniskacniska
View sourceSee grades on GitHubMore from cniska →
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

Competitor Analysis

This skill provides comprehensive analysis of competitor SEO and GEO strategies, revealing what's working in your market and identifying opportunities to outperform the competition.

1823 votes

Deep Research

Universal deep research agent team. 13-agent pipeline for rigorous academic research on any topic. 8 modes: full research, quick brief, paper review, lit-review, fact-check, three-way literature scan, Socratic guided research dialogue, and systematic review with optional meta-analysis. Covers research question formulation, Socratic mentoring, methodology design, systematic literature search, source verification, cross-source synthesis, risk of bias assessment, meta-analysis, APA 7.0 report co...

502942 votes

Paperclip Distill

Use when an operation issue is a Paperclip cursor-window, distill, or backfill — `operationType: "distill"` or `"backfill"` and the body references a Paperclip source bundle for a project or root issue. Turn raw Paperclip activity into a wiki-insightful project page, decisions log, and history note. This skill exists specifically to replace the stiff, datestamp-heavy templated output that the deterministic distiller produces.

953191 votes

Academic Pipeline

Orchestrator for the full academic research pipeline: research -> write -> integrity check -> review -> revise -> re-review -> re-revise -> final integrity check -> finalize. Coordinates deep-research, academic-paper, and academic-paper-reviewer into a seamless 10-stage workflow with mandatory, coverage-bounded integrity checks, two-stage peer review, and auditable quality-assurance artifacts. Triggers on: academic pipeline, research to paper, full paper workflow, paper pipeline, end-to-end p...

502941 votes

Literature Review

Assistance with writing literature reviews by searching for academic sources via Semantic Scholar, OpenAlex, Crossref and PubMed APIs. Use when the user needs to find papers on a topic, get details for specific DOIs, or draft sections of a literature review with proper citations.

6511 votes
View all in research →