spec-authoring
Use this skill to propose changes via the Spec PR process. It uses the Gemini CLI to generate high-quality draft specifications and to analyze PR feedback, accelerating the spec-driven development workflow. Triggers include "create spec" or "propose change".
What this skill does
# Spec Authoring Skill
## Purpose
To manage the creation and refinement of feature specifications using a powerful, AI-assisted workflow. This skill leverages the **Gemini CLI** to accelerate the spec-driven development process by:
1. **Generating Drafts**: Automatically creates high-quality, multi-file draft proposals for new features.
2. **Analyzing Feedback**: Synthesizes review comments from Pull Requests into an actionable summary of recommended changes.
This approach allows developers and product managers to move from idea to an approved, implementation-ready specification with greater speed and clarity.
## When to Use
Use this skill in the following situations:
- Proposing a new feature or significant change.
- Generating a first draft of a specification for review.
- Processing and incorporating feedback from a Spec PR.
## Prerequisites
- Project initialized with AgenticDev structure (`docs/specs/` and `docs/changes/` directories exist).
- GitHub repository set up.
- `gh` CLI tool installed and authenticated.
- `gemini` CLI tool installed and authenticated.
## Spec PR Philosophy
**Specs as Code**: All specification changes follow the same rigor as code changes—proposed via branches, reviewed via PRs, and merged upon approval. This skill supercharges that philosophy with AI.
---
## The `propose` Command
### Purpose
Generate a comprehensive, multi-file draft proposal for a new feature from a single command.
### Workflow
#### Step 1: Define the Proposal Name
Choose a clear, descriptive name for your feature, such as "User Authentication System" or "Real-time Notifications".
#### Step 2: Run the Helper Script
Execute the script to generate the draft proposal:
```bash
bash scripts/spec-authoring.sh propose "Feature Name"
```
The script will:
1. Create a new directory in `docs/changes/feature-name/`.
2. Make three parallel calls to the **Gemini CLI** to generate drafts for `proposal.md`, `spec-delta.md`, and `tasks.md`.
- **`proposal.md`**: A high-level overview with problem statement, proposed solution, and success criteria.
- **`spec-delta.md`**: A detailed technical specification with requirements and design decisions.
- **`tasks.md`**: A preliminary breakdown of implementation tasks.
3. Save the AI-generated content into these files.
#### Step 3: Review and Refine the Drafts
The script provides you with a complete, context-aware first draft of your entire proposal. Your next step is to review and refine these documents to ensure they align with your vision before opening a Spec PR.
---
## The `update` Command
### Purpose
Intelligently process feedback from a Spec PR by using AI to analyze review comments and generate a summarized action plan.
### Workflow
#### Step 1: Identify the PR Number
Determine which Spec PR you need to update.
#### Step 2: Run the Feedback Analysis Script
Execute the script with the PR number:
```bash
bash scripts/spec-authoring.sh update PR_NUMBER
```
This command will:
1. Find the local files associated with the PR's branch.
2. Fetch all review comments from the PR.
3. Send the full content of your spec files and all the comments to the **Gemini CLI**.
4. Ask the AI to act as a reviewer and provide a summarized list of recommended changes for each file.
#### Step 3: Address the Synthesized Feedback
The script will output a clear, actionable plan that synthesizes all the reviewer feedback. Use this analysis to efficiently update your proposal files, address the comments, and push your changes for re-review.
---
## Error Handling
### Gemini CLI Issues
**Symptom**: The script fails during the `propose` or `update` commands with an error related to the `gemini` command.
**Solution**:
- Ensure the `gemini` CLI is installed and in your system's PATH.
- Verify you are authenticated (`gemini auth`).
- Check for Gemini API outages or network issues.
### Proposal Directory Already Exists
**Symptom**: The `propose` command reports that the directory already exists.
**Solution**: Choose a different name for your proposal or work with the existing one.
### Could Not Find Proposal Directory
**Symptom**: The `update` command cannot find the local files for the PR.
**Solution**: Ensure you have the correct PR branch checked out and that the local directory in `docs/changes/` matches the branch name.
## Notes
- **AI-Assisted Workflow**: This skill is designed to be a powerful assistant. It generates high-quality drafts and analyzes feedback, but the final strategic decisions and refinements are yours to make.
- **Speed and Quality**: By automating the initial drafting and feedback synthesis, this skill allows you to focus on the high-value work of design, review, and alignment.
- **Iterative Process**: Use the `propose` command to start, and the `update` command to iterate based on team feedback, creating a rapid and efficient spec development cycle.Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.