How do I install the Stackweft Delivery skill?

Run npx skillmds@latest add loping151/stackweft-delivery in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Stackweft Delivery safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Stackweft Delivery?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Stackweft Delivery free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Stackweft Delivery?

loping151 (@loping151) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Stackweft Delivery by loping151 — Agent Skill

Name: Stackweft Delivery
Author: loping151

Stackweft Delivery

Delegate full-stack field/feature delivery to StackWeft — a compiled, evidence-gated pipeline for React + Express + Sequelize repos. Use when the user wants to ADD or EXTEND a cross-stack field (Sequelize model → controller / API payload → frontend service → list-card AND detail-page DOM) and have it verified by counterexample sentinel probes. Cheaper and steadier than doing it yourself in a free agentic loop, and it remembers repeated work (0-LLM replay). Do NOT use for non-field work (algorithms, infra, cross-service refactors) or repos that don't fit this full-stack field-flow shape — handle those yourself.

loping151 62ecd11 2.5 KB Updated Sep 17, 2026

File contents

StackWeft delivery (delegated subagent)

You are the host agent. For a cross-stack field/feature delivery in a real full-stack repo, do not hand-edit across the stack yourself — delegate the whole job to StackWeft and report back its verified result.

Set STACKWEFT_HOME to this repo's absolute path (or edit the paths below).

Steps

Delegate. Pass the user's requirement verbatim (natural language is fine):
```
"$STACKWEFT_HOME/sw" run "<the user's requirement>" --repo "<abs repo path>"
```
- --repo is optional; without it StackWeft resolves the workspace from the requirement text (and initializes a git baseline for non-git folders).
- The command prints run_id=… near the top. Exit code 0 = all gates passed, 1 = not passed.
Collect the structured result and base your report on it (do not re-verify by hand — the probes already did):
```
"$STACKWEFT_HOME/sw" json --debug
```
This emits one JSON object: stages (clarify→plan→localize→compile→generate→ verify→pr), per-call tokens, pass/fail per gate, the branch, and PR info.
Report to the user: what field landed, that the sentinel probes are green (value reaches model + API payload + list/detail DOM), the branch/PR, and the token cost. If exit code was 1, surface which gate failed from the JSON instead of claiming success.

Run-time control (optional)

A delivery is pausable/resumable; you generally don't need this, but:

"$STACKWEFT_HOME/sw" control <run_id> pause|abort|append|resume ["append text"]

Don't

Don't fabricate a run_id, token count, or "passed" — only report values you actually read from sw json this turn.
Don't fall back to editing files yourself unless StackWeft returns failure AND the user asks you to take over.

---name: stackweft-deliverydescription: Delegate full-stack field/feature delivery to StackWeft — a compiled, evidence-gated pipeline for React + Express + Sequelize repos. Use when the user wants to ADD or EXTEND a cross-stack field (Sequelize model → controller / API payload → frontend service → list-card AND detail-page DOM) and have it verified by counterexample sentinel probes. Cheaper and steadier than doing it yourself in a free agentic loop, and it remembers repeated work (0-LLM replay). Do NOT use for non-field work (algorithms, infra, cross-service refactors) or repos that don't fit this full-stack field-flow shape — handle those yourself.---# StackWeft delivery (delegated subagent)You are the host agent. For a cross-stack **field/feature delivery** in a realfull-stack repo, do not hand-edit across the stack yourself — delegate the wholejob to StackWeft and report back its verified result.Set `STACKWEFT_HOME` to this repo's absolute path (or edit the paths below).## Steps1. **Delegate.** Pass the user's requirement verbatim (natural language is fine): ```bash "$STACKWEFT_HOME/sw" run "<the user's requirement>" --repo "<abs repo path>" ``` - `--repo` is optional; without it StackWeft resolves the workspace from the requirement text (and initializes a git baseline for non-git folders). - The command prints `run_id=…` near the top. Exit code **0 = all gates passed**, **1 = not passed**.2. **Collect the structured result** and base your report on it (do not re-verify by hand — the probes already did): ```bash "$STACKWEFT_HOME/sw" json --debug ``` This emits one JSON object: stages (clarify→plan→localize→compile→generate→ verify→pr), per-call tokens, pass/fail per gate, the branch, and PR info.3. **Report** to the user: what field landed, that the sentinel probes are green (value reaches model + API payload + list/detail DOM), the branch/PR, and the token cost. If exit code was 1, surface which gate failed from the JSON instead of claiming success.## Run-time control (optional)A delivery is pausable/resumable; you generally don't need this, but:```bash"$STACKWEFT_HOME/sw" control <run_id> pause|abort|append|resume ["append text"]```## Don't- Don't fabricate a `run_id`, token count, or "passed" — only report values you actually read from `sw json` this turn.- Don't fall back to editing files yourself unless StackWeft returns failure AND the user asks you to take over.

Frequently asked questions