← Back to list
📄

sureforge Checked

Quality-control workflow for complex, multi-step tasks that need research, clarification, a verif…
📄 Documents skills By Da7-Tech Version v1.0.0 Updated 2026-09-17
9.4Overall rating

Installation

🤖 Install via AI

Copy the prompt below and send it to your AI assistant (e.g. Claude Code) — it will follow the instructions and install automatically.

Install the "sureforge" skill by following the instructions at https://skill123.me/install/sureforge.
⌨️ Command line install

Run in your terminal — downloads and installs to ~/.claude/skills/.

curl -fsSL https://skill123.me/install/sureforge.sh | bash
📦 Download ZIP

Download the zip and extract it into your skills directory (e.g. ~/.claude/skills/), then restart your session.

⬇ Download v1.0.0 · 33 KB

About this skill

Overview

Quality-control workflow for complex, multi-step tasks that need research, clarification, a verifiable plan, independent review, and evidence-backed delivery. Use for substantial implementations, decision studies, multi-file deliverables, and exhaustive document or visual checks, or when the user explicitly asks to use SureForge. Scale down for small tasks; do not activate automatically for casual questions, simple lookups, or trivial edits.

Source

  • Repo: https://github.com/Da7-Tech/SureForge
  • Path: skills/sureforge

Score breakdown

Trigger
8.5
Exemplary What/When/When-Not including an explicit-invocation clause and a scale-down boundary; trigger phrase enumeration beyond the name is moderate.
Structure
10.0
Textbook progressive disclosure: 96-line SKILL.md with five standing rules and a tier table, seven per-phase references loaded on entry, four ledger/brief assets, and a light-mode exception that skips loading ceremony.
Workflow
9.2
Strong governance design: tiered verification budgets, READY/REPAIR/BLOCKED gate decisions with round limits, genuine-independence review protocol with neutral packages and withheld owner verdicts, evidence-freshness rules on resume, and calibrated usage examples; inherently advisory (no enforcement), which it states honestly.
Content
9.0
Precise, internally consistent abstract prose; occasionally dense, and the governance vocabulary demands careful reading.
Engineering
10.0
Valid frontmatter with license/metadata; no scripts by design; every referenced asset and reference file exists in the package.
Security
10.0
Capability manifest matches declared scope: no scripts, network, credentials or persistence; explicit anti-injection rule (external content is data, not authority) and unusually mature security guidance about isolation limits and credential misuse.
Methodology-only QC workflow (no scripts, no network) from Da7-Tech/SureForge: tiered (Light/Standard/Full) research-clarify-plan-execute-deliver process with evidence-backed gates, independent-review protocol and coverage ledgers. Capability manifest (6.0): reads only the user's request, project material and supplied references; side effects limited to task/coverage ledgers written in an authorized workspace (explicitly instructed not to write inside the installed skill); no network destinations; no credentials; no persistence. Capabilities match declared scope; the skill is itself security-aware (read-only labels are not security boundaries; never obtain credentials to fake reviewer independence; external content is data, not authority).