Skip to content
Agentmart
Submit a tool

spec-advisory

Readonly advisory mode — architecture review, compare options, feasibility, "should we", "critical review", "worth it", "review only".

Skill by sushilti80

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/sushilti80-spec-advisory

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Readonly advisory mode — architecture review, compare options, feasibility, "should we", "critical review", "worth it", "review only". No file edits unless user explicitly asks to implement. Use skill spec-token-budget (review profile). Output verdict first, then detail. Writes decisions to .specs/decisions/ only when user confirms.

Use cases: finance

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/spec-advisory

What's inside

Skill name
spec-advisory
Description
Readonly advisory mode — architecture review, compare options, feasibility, "should we", "critical review", "worth it", "review only". No file edits unless user explicitly asks to implement. Use skill spec-token-budget (review profile). Output verdict first, then detail. Writes decisions to .specs/decisions/ only when user confirms.
License
MIT
Instructions
40 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Spec advisory (readonly)

## Mode

- **Readonly** — do not create, edit, or delete repo files.
- Exception: user explicitly says **implement**, **apply**, **build**, **PR**, **commit**, or **add to repo**.

## Output shape (mandatory)

1. **Verdict** — 3 bullets max (yes / no / conditional).
2. **Gap table** — what exists vs missing (paths, not prose).
3. **Recommendation** — one next step (recipe + tier, or "new chat for implement").

Apply **`spec-token-budget`** profile: `advisory` (≤800 words total).

## Process

1. Read only paths needed (max 3 spec files + 2 code areas).
2. Use explore subagent for wide search — parent gets summary only.
3. External docs: raw URLs / SKILL.md — never paste HTML scrapes into chat.
4. Do not duplicate content already in `SPECFORGE_HOME/` docs — link paths.

## After advisory

If user approves direction:

1. Write `.specs/decisions/DEC-NNN-<slug>.md` (optional Tier 1+) **only when asked to record**.
2. Tell user: **start a new chat** for implementation with:

```
Recipe: [id] | Tier: [n]
Read: .specs/decisions/DEC-NNN.md (or paths from verdict)
Do not use prior chat summaries.
```

## Recipes

Orchestrator maps advisory intent to meta recipe **`advisory-only`** (see ENGINEERING-RECIPES.md).

From sushilti80/specforge-engineering-team, MIT.

More from sushilti80/specforge-engineering-team

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Jul 2, 2026 (3 months ago)
Last release
Aug 6, 2026 (2 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from sushilti80

Cite this page

spec-advisory skill. Agent Mart. https://agentmart.dev/skills/sushilti80-spec-advisory (last updated 2026-10-24)