bingordinary/specflow/using-specflow-guidance

Claude Code and agent skill by bingordinary/specflow

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Use when a natural-language specFlow request may need product or design guidance before formal truth writeback, especially vague project ideas, scope choices, solution options, or design review before candidate Spec editing.

Install

npx vskill@latest install bingordinary/specflow/using-specflow-guidance

SKILL.md

Using specFlow Guidance

Purpose

Use this skill to decide whether a natural-language request should enter a guidance skill before a formal specFlow command or truth writeback.

Guidance helps a user form a better project design. It does not become durable truth.

Routing Rule

Use guidance when the request is about any of these before the formal candidate truth is clear:

  1. framing a vague project or feature idea
  2. cutting scope for a first useful version
  3. choosing between materially different solution directions
  4. reviewing a design before writing it into candidate truth
  5. turning an approved discussion result into candidate truth

Before selecting a guidance skill, diagnose the user's goal in ordinary language. Do not ask the user to classify the request as unit, rule, or stable g_ rule. Use repository truth and natural-language routing to translate the user's goal into the internal specFlow owner.

Skill Selection

  1. Use project-framing when the goal, user, problem, or success meaning is unclear.
  2. Use scope-cutting when the request is too broad for one candidate round or mixes several independent capabilities.
  3. Use solution-design when the goal is clear but the solution direction is not locked.
  4. Use design-quality-review when a discussion-stage design is about to be written into candidate truth.
  5. Use spec-writeback-guidance when the user has approved a design conclusion that must become durable candidate truth.

Hard Boundaries

  1. Do not implement from guidance output.
  2. Do not treat chat-only agreement as durable truth.
  3. Once a conclusion affects behavior, boundary, acceptance, rule truth, or global rule truth, route it into formal specFlow truth writeback before implementation.
  4. The legality basis for every route, target, and next step in these skills is framework/core/object_model.md (object family and layer rules), framework/spec_writing_guide.md (path, writeback, and rule-scope rules — §10 for rule scope resolution), and the candidate-to-promote workflow in framework/concepts.md.
  5. Do not turn guidance into a command-alias conversation where the user must choose internal command or object-family names.
  6. Ask only ordinary-language questions about goal, scope, outcome, success meaning, or required human judgment.

Completion

A guidance step ends with one of these outcomes:

  1. continue_discussion - the design is not clear enough yet
  2. ready_for_candidate_writeback - the user has approved a specific conclusion that should be written into candidate truth
  3. route_to_existing_command - current repository truth already provides a legal command route
  4. do_not_writeback - the guidance result is a blocking design issue that must not be encoded into candidate truth (returned by design-quality-review); the step ends here — no writeback is offered, and the blocking issue is reported to the user with the writeback_ready_points and points_to_exclude_from_candidate lists
  5. stop_for_checkpoint - the smallest missing decision must be requested through the active routing or command checkpoint rules

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/bingordinary/specflow/using-specflow-guidance/badge)](https://verified-skill.com/skills/bingordinary/specflow/using-specflow-guidance)

Something wrong with this skill or its page? Tell us.