# Decomplex

> >-

- **Type:** Skill
- **Install:** `agentstack add skill-maxedapps-agent-skills-decomplex`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [maxedapps](https://agentstack.voostack.com/s/maxedapps)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [maxedapps](https://github.com/maxedapps)
- **Source:** https://github.com/maxedapps/agent-skills/tree/main/skills/decomplex

## Install

```sh
agentstack add skill-maxedapps-agent-skills-decomplex
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Decomplex

## Hard rules

- Exactly one mode:
  - **Prevention** — unimplemented plans/designs/deps/safeguards
  - **Audit** — existing source/tests/config/deps/architecture
  - **Finding triage** — supplied recs that may add complexity
- Fix before reviewing:
  - target
  - authority
  - required behavior
  - scope
- Ask only when ambiguity changes those.
- Targets immutable.
- **Only** write: one `.reviews/-decomplex.md`.
- Never edit reviewed artifacts (even if asked to fix).
- Direct invoke and report not writable → **stop**.
  - no chat-only substitute
  - embedders may use their own fallback
  - must not claim `decomplex` ran
- Complexity trade-offs only.
- Route defects/security/perf/plan-compliance to the owning workflow (e.g. `code-review`).
- Simplicity ≠ fewer lines/helpers/deps.
- Preserve evidenced:
  - trust boundaries
  - invariants
  - compatibility
  - lifecycle cleanup
  - ops needs
- Findings are advisory.
- Consumer dispositions/implements/escalates.
- **“No potential complexity findings” is valid.**
- Consumers must:
  - disposition every item
  - never auto-implement
  - **ask user** when in doubt

## Admission gate (all required)

1. Concrete conceptual, maintenance, or operational burden
2. Absent/insufficient justification for current need
3. Realistic, reachable practical cost
4. Smallest concrete simpler alternative
5. Required behavior preserved
6. Proportionate value after simplification + regression cost
7. Explicit exception/boundary check

Evidence:

- `Confirmed` or `Supported` only

If the gate fails:

- bounded validation
- user question
- or omit

Never find on:

- helper count
- duplication alone
- novelty
- hypothetical scale
- style alone

Prefer:

1. delete
2. direct use
3. local logic/helper
4. existing shared abstraction
5. new abstraction

## Sequence

1. **Contract**
   - mode
   - target
   - authority
   - scope
   - behavior
   - report path
   - constraints
   - inspect callers/tests/config/history only as needed for the gate
2. **Before candidates**
   - read [`references/complexity-gates.md`](references/complexity-gates.md) fully
   - contextual signals, not a quota checklist
3. **Evaluate**
   - full gate each candidate
   - order by expected simplification value
4. **Recommend**
   - only: `Act` | `Validate` | `Ask user`
   - do not perform the action
   - triage mode: preserve every supplied ID
   - triage dispositions: `Act` | `Validate` | `Ask user` | `No action`
5. **Before write**
   - read [`assets/decomplex-review-template.md`](assets/decomplex-review-template.md) fully
   - write exactly one report
   - strip template guidance/empty sections
   - keep inspected/skipped coverage + limitations
6. **Verify**
   - report exists
   - admitted items pass gate
   - all triage IDs accounted
   - no target file changed

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [maxedapps](https://github.com/maxedapps)
- **Source:** [maxedapps/agent-skills](https://github.com/maxedapps/agent-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-maxedapps-agent-skills-decomplex
- Seller: https://agentstack.voostack.com/s/maxedapps
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.