菜单
Generates a self-contained prompt pack from a spec. Embeds real code patterns from .vibeflow/ so any coding agent follows the project's conventions. Use when handing off implementation to a separate session or agent.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Audits recent work against its Definition of Done and project patterns. Runs the test suite, compares code against the spec, and reports PASS / PARTIAL / FAIL. Also runs the Critical Gate — a safety scan of the diff for destructive or dangerous operations. Generates an incremental prompt pack for any gaps found. Use after implementation to verify quality before shipping.
Audits implementation against its DoD and project patterns. Runs the test suite, compares code against the spec, and reports PASS / PARTIAL / FAIL. Also runs the Critical Gate — a safety scan of the diff for destructive or dangerous operations. Use after implementation to verify quality before shipping.
Implements a feature from its spec following all guardrails: budget, DoD, anti-scope, and pattern compliance. Runs an 8-phase pipeline (find spec → extract guardrails → load patterns → plan → implement → test → refine → self-verify DoD). Use after gen-spec when you're ready to code. The agent has filesystem access and acts as Coding Agent.
Implements a feature from its spec with guardrails: budget, DoD, anti-scope, and pattern compliance. Runs an 8-phase pipeline (find spec → extract guardrails → load patterns → plan → implement → test → refine → self-verify DoD). Use after gen-spec when ready to implement.
Deep-analyzes the current codebase to discover stack, architecture, patterns, conventions, and pitfalls. Creates curated documentation in .vibeflow/ that persists and can be committed to git. Supports incremental updates, scoped deep-dives, interactive review, and satellite repo analysis. Use when setting up a project, after significant changes, or to deep-dive into a specific module.
Generates a technical spec with objective, Definition of Done (3–7 binary checks), scope, anti-scope, technical decisions, and applicable patterns. Grounded in the project's real patterns from .vibeflow/. Auto-splits specs that exceed budget or DoD limits. Use when requirements are clear and you're ready to define the implementation contract.
| name | vibeflow-prompt-pack |
| description | Generates a self-contained prompt pack from a spec. Embeds real code patterns from .vibeflow/ so any coding agent follows the project's conventions. Use when handing off implementation to a separate session or agent. |
Generate a self-contained prompt pack from a spec. The prompt pack
is designed for a coding agent (Copilot, Cursor, Claude Code, etc.) that
has NO context beyond the prompt itself. Embeds real patterns from
.vibeflow/ so the agent follows the project's conventions.
Usage: Provide the spec file path or feature name as input.
The prompt pack MUST be written in the user's detected language. The opening line should be language-adaptive: Write the opening line in the user's detected language. The concept is: "You are only seeing this prompt; there is no context outside it."
All textual content (objective, DoD, anti-scope, guidance) should be in the detected language. Code, paths and technical names remain in English.
Validate spec size. Read the spec and check:
.vibeflow/index.md or default ≤ 6 files)If EITHER condition is true, stop. Do NOT generate the prompt pack. Instead, inform the user: "This spec exceeds limits (N DoD checks / N files). Use the vibeflow-gen-spec skill again to split it into smaller specs before generating a prompt pack."
If the input is a file path, read that spec file.
If it's a feature description, look for a matching spec in .vibeflow/specs/.
If no spec exists, generate one first (following gen-spec format), save it, then continue.
Read .vibeflow/ docs:
.vibeflow/conventions.md (always)## Pattern Registry YAML block from index.md (between
<!-- vibeflow:patterns:start/end --> markers). Cross-reference the
registry's tags and modules against the spec's scope. Load the top 3-5
matching patterns. If no registry exists, infer which ones are relevant.Read the actual codebase files relevant to this task.
Generate the prompt pack.
You are only seeing this prompt; there is no context outside it.
(Write this opening line in the user's detected language.)
Copied from the spec. Non-negotiable.
What NOT to do. Copied from spec.
Max files to change. Read the budget from the spec. If the spec doesn't
specify a budget, check .vibeflow/index.md for the Suggested budget
line. If neither is available, default to ≤ 6.
THIS IS CRITICAL. Embed the actual patterns the agent must follow.
Copy the relevant sections from .vibeflow/patterns/*.md and
.vibeflow/conventions.md directly into the prompt pack.
Include REAL code examples from the pattern docs — not references to external files the agent won't see. The prompt pack is self-contained.
Format as:
## Patterns to Follow
### <Pattern Name>
<description>
<real code example from this project showing the pattern>
### Coding Conventions
<relevant conventions for this task>
Real file paths from the codebase (verify they exist). Relevant code snippets for context.
Architectural direction, constraints to respect. NOT step-by-step prescriptive instructions.
Commands to validate the implementation. This section is REQUIRED.
.vibeflow/index.md or project config files (package.json,
pyproject.toml, Cargo.toml, etc.) to detect the test runner.npm test, pytest,
cargo test, go test ./...).## How to validate
1. Run tests: `npm test`
2. Verify manually: [description]
Which docs need changes after implementation.
If you cannot verify a file path exists, flag it with:
<!-- TODO: verify this path -->
Save the prompt pack to: .vibeflow/prompt-packs/<feature-slug>.md
Create the .vibeflow/prompt-packs/ directory if it doesn't exist.
After saving, suggest: "Prompt pack saved. Hand it off to the coding agent. After implementation, use the vibeflow-audit skill with the spec to verify."