> For the complete documentation index, see [llms.txt](https://docs.walnutai.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.walnutai.ai/core-features/gap-analysis/phase-2-missing-story-detection.md).
# Phase 2: Missing Story Detection
**Phase 2:** Missing Story Detection identifies functionality that exists in the codebase but is not documented as user stories. WalnutAI also validates existing stories to assess coverage and provides AI-driven suggestions for improving and updating them. In this phase, the code acts as the baseline, and WalnutAI detects undocumented or incomplete requirement coverage.
* **Before starting the analysis:**
* The project must be connected to one or more repositories.
* Ensure the repositories were properly integrated during project creation.
* The codebase from the selected repositories will be used as the reference for comparison against existing user stories.
* **To start the analysis:**
* Click **Start Analysis**
* Select one or more connected repositories.
* Click **Start Analysing** to begin evaluation.
* **During analysis:**
* WalnutAI scans the selected repository and branch.
* It compares implemented code with documented user stories.
* It identifies functionality implemented in code but not documented as a story.
* It detects stories that require corrections or acceptance criteria improvements.
* It enhances acceptance criteria to fully reflect implemented behaviour.
* It generates new user story suggestions for uncovered features.
* **After the analysis completes, the dashboard displays:**
* **Stories Analysed** – Total number of stories reviewed.
* **Corrections Needed** – Stories requiring refinement.
* **Missing Stories** – Undocumented features detected from code.
* **High Priority** – Critical missing or impacted stories.
* **In the Analysis Results section:**
* Separate tabs for **Corrected Stories** and **Missing Stories** with respective counts.
* **For Corrected Stories:**
* Click on Option to **Select All** or select individual stories and directly **Apply Correction** using the **Apply Correction** action.
* Clear selection option to reset selections.
* Click on Story for Side-by-side comparison of **Original** and **Corrected** versions.
* Updated Title, Description, and Acceptance Criteria suggested by Walnut AI.
* Highlighted improvements in acceptance criteria coverage.
* Confidence percentage indicating alignment strength.
* **AI Reasoning** explaining why corrections are recommended.
* **For Missing Stories:**
* Click on Option to **Select All** or select individual stories and directly create user stories using the **Create Story** action.
* Clear selection option to reset selections.
* Suggested story title derived from code behaviour.
* Priority level (e.g., High).
* The **Confidence Score** indicates how strongly the detected functionality aligns with the suggested or corrected user story based on Walnut AI’s semantic and structural analysis.
* **Functionality Description** generated from implementation.
* **Suggested Acceptance Criteria** based on detected logic.
* **Related Code Files** showing impacted source files.
* Click **Run New Analysis** to scan the updated codebase and user stories again, ensuring newly implemented features and recent changes are reflected in the analysis results
Phase 2 ensures implemented features are fully documented, improves requirement and acceptance criteria quality, maintains code-to-requirement traceability, and keeps documentation aligned with actual system behaviour while preventing undocumented development.
