Install
$ agentstack add skill-dennisle-lts-claude-skills-sdlc-doc-generator ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
SDLC Document Generator
Generates the following documents in Markdown (.md) format from source project information:
- BRD — Business Requirements Document
- PRD — Product Requirements Document
- TDD — Technical Design Document
- Deployment Plan
- Release Notes
- ADR — Architecture Decision Record
Workflow
Phase 1 — Source Information Gathering
- Ask the user to provide the folder path that contains any existing project information. This could include:
- Existing docs (Word, PDF, MD, TXT)
- Meeting notes
- Email exports
- Spreadsheets
- Wireframes descriptions
- Any other reference material
- Read all files in the folder using
bash_tool:
``bash ls -la /path/to/folder cat /path/to/folder/*.md # or txt, etc. ``
- Analyze the content and map what you've found to the required sections of each document.
- Build an internal information gap list — what is missing or unclear per document.
Phase 2 — Follow-up Questions
After reading the folder, ask targeted follow-up questions only for information that is missing and cannot be reasonably inferred.
Group questions by theme to avoid overwhelming the user. Example groupings:
- Business context (for BRD gaps)
- Product & user details (for PRD gaps)
- Technical decisions (for TDD & ADR gaps)
- Release & deployment details (for Deployment Plan & Release Notes gaps)
Important rules:
- Never ask for information already present in the source files
- Ask a maximum of 10 questions at a time
- Prefer multiple choice or short answers where possible
- If a detail is minor and can be inferred, make a reasonable assumption and note it in the document
Phase 3 — Document Generation
Generate each requested document as a separate .md file. Save all files to the same output folder as the source, or to /mnt/user-data/outputs/ if no writable folder is available.
File naming convention:
BRD_.md
PRD_.md
TDD_.md
DeploymentPlan_.md
ReleaseNotes_.md
ADR__.md
For each document, read the corresponding template from the references/ folder before generating:
references/brd-template.mdreferences/prd-template.mdreferences/tdd-template.mdreferences/deployment-plan-template.mdreferences/release-notes-template.mdreferences/adr-template.md
Phase 4 — Review & Iteration
After generating all documents:
- Present the files to the user using
present_files - Ask: "Would you like to refine any section or add more detail to a specific document?"
- Apply changes and re-present updated files
Key Principles
- Never fabricate critical details like version numbers, dates, stakeholder names, or technical decisions — ask if missing
- Always note assumptions made in an
> ℹ️ Assumption:callout block inline in the document - Keep language consistent with the source material (match formality level)
- BRD uses business language — avoid technical jargon
- TDD and ADR use technical language — be precise and specific
- Release Notes are concise — use bullet points, avoid paragraphs
- For ADR, generate one ADR file per architectural decision identified
Document Selection
If the user has not specified which documents they want, ask:
> "Which documents would you like me to generate?" > - BRD > - PRD > - TDD > - Deployment Plan > - Release Notes > - ADR > - All of the above
Default to All if the user says "all project docs" or similar.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: dennisle-lts
- Source: dennisle-lts/claude_skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.