Speckit Baseline

Speckit Baseline Skill

The user input to you can be provided directly by the agent or as a command argument - you MUST consider it before proceeding with the prompt (if not empty).

User input:

$ARGUMENTS

What This Does

The baseline workflow creates foundational documentation by:

1. Establishing a comprehensive baseline specification of the project
2. Creating a change tracking document for all subsequent modifications

This provides essential context for you (the AI agent) when working on future tasks.

Your Task

1. Run the script .specify/scripts/bash/create-baseline.sh --json from repo root and parse its JSON output for BASELINESPEC, CURRENTSTATE, BASELINECOMMIT, and HASEXISTING_SPECS. All file paths must be absolute.

Note: The script only accepts the --json flag and ignores $ARGUMENTS. IMPORTANT You must only ever run this script once. The JSON is provided in the terminal as output - always refer to it to get the actual content you're looking for.

1. Analyze the Project:

a. If HASEXISTINGSPECS is "true":

• The baseline commit represents the project state BEFORE spec-kit was used
• Use git show BASELINE_COMMIT:path/to/file to examine files at that point
• Analyze that historical state to document what existed before specs

b. If HASEXISTINGSPECS is "false":

• Analyze the current project state
• This is a fresh baseline for future work

1. Generate baseline-spec.md:

Load .specify/extensions/workflows/baseline/baseline-spec-template.md to understand the structure.

Then comprehensively analyze the codebase (at the baseline commit if applicable) and fill in BASELINE_SPEC with:

• Executive Summary: High-level overview of project purpose
• Project Structure: Directory layout and key components
• Architecture: System design, technology stack, patterns
• Core Functionality: Primary features and user workflows
• Data Model: Core entities and their relationships
• External Dependencies: Third-party libraries and external services
• Build and Deployment: Build process, testing strategy, deployment
• Configuration: Environment variables, feature flags
• Known Issues: Current limitations and technical debt
• Future Considerations: Planned improvements and scalability concerns

Be thorough - examine:

• Package files (package.json, requirements.txt, pom.xml, etc.)
• Build configurations (webpack, vite, Makefile, etc.)
• Source code structure and key modules
• README, documentation files
• Configuration files
• Database schemas or migrations
• API routes/endpoints
• Test files to understand tested functionality

1. Generate current-state.md:

Load .specify/extensions/workflows/baseline/current-state-template.md for structure.

Scan the specs directory and enumerate all changes by type:

• Features: List all specs/###-* directories with descriptions
• Modifications: Find specs/###-*/modifications/ subdirectories
• Bugfixes: List specs/bugfix/ entries
• Refactors: List specs/refactor/ entries
• Hotfixes: List specs/hotfix/ entries
• Deprecations: List specs/deprecate/ entries

For unspecified changes (if HASEXISTINGSPECS is true):

• Compare commits from BASELINE_COMMIT to HEAD
• Identify commits not associated with spec directories

Calculate statistics:

• Counts for each workflow type
• Total commits, files changed, lines added/removed
• Specification coverage percentage

1. Report completion:

✅ Baseline documentation created

**Location**: specs/history/
**Baseline Spec**: [BASELINE_SPEC path]
**Current State**: [CURRENT_STATE path]

{{If HAS_EXISTING_SPECS is true:}}
**Baseline Commit**: [BASELINE_COMMIT]
Note: Baseline represents project state before first spec-kit usage
{{end if}}

📋 **What Was Documented:**

**Baseline Specification**:
- Project structure: [X] components identified
- Technology stack: [Languages/Frameworks]
- Core features: [X] primary features documented
- Architecture: [Brief architecture summary]

**Current State**:
- Features: [X] implemented
- Modifications: [X] tracked
- Bugfixes: [X] fixed
- Other workflows: [X] total
{{If unspecified changes found}}
- Unspecified: [X] commits without specs
{{end if}}

💡 **Using This Context:**

The baseline documentation will help me (your AI assistant) understand:
- What the project does and how it's structured
- What has been built and modified
- Where technical debt exists
- What patterns and conventions to follow

This context will improve my suggestions for all future work.

📌 **Next Steps:**
1. Review the generated documentation for accuracy
2. Make any corrections or additions to baseline-spec.md
3. Use this context when creating new specs with other workflows
4. Periodically update current-state.md as new specs are completed

Important Notes

• Do not create a git branch - This is a documentation task
• Be comprehensive - The baseline is foundational context for all future work
• Analyze deeply - Don't just list files, understand what they do
• Track everything - In current-state.md, account for all changes
• Update timestamps - Replace {{DATE}} placeholders with actual dates
• Link to sources - Reference specific files and line numbers where relevant

Quality Checks

Before finishing:

• [ ] baseline-spec.md covers all major aspects of the project
• [ ] Technology stack is accurately documented
• [ ] Core features are clearly described
• [ ] current-state.md accounts for all spec directories
• [ ] Statistics are calculated and included
• [ ] Both files use the provided template structure
• [ ] All placeholders are replaced with actual content

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: pradeepmouli
• Source: pradeepmouli/skillit
• License: MIT
• Homepage: https://pradeepmouli.github.io/skillit/

Install and usage instructions live in the source repository linked above.