project-memory

Recall prior project decisions and patterns via tag-overlap search with grep fallback.

Updated Apr 27, 2026
One-click install
npx skills add https://github.com/vTRKA/supervibe --skill project-memory-vtrka
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: project-memory
Source: https://github.com/vTRKA/supervibe/tree/main/skills/project-memory
Command: npx skills add https://github.com/vTRKA/supervibe --skill project-memory-vtrka

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Project Memory prevents wasted effort and risky repetition by helping the AI recall earlier decisions, patterns, incidents, and solutions that match the current task context before starting non-trivial work.

Core Features & Use Cases

  • Memory-backed brainstorming: Searches .supervibe/memory/ using tag overlap first, then falls back to keyword/grep search when needed.
  • Evidence-aware synthesis: Reads relevant entries, applies confidence and staleness checks, and turns matches into an actionable recommendation for the requesting agent.
  • Safety guardrails: Blocks when required memory artifacts, runtime proof, or expected inputs are missing, instead of guessing or approving durable work without evidence.
  • Decision, pattern, incident, learning, and solution recall: Consolidates multiple memory types into a bounded handoff (up to five entries) with citations to file paths.

Quick Start

Ask your AI to look up project memory for prior decisions relevant to the task you are about to start, and then use the returned matches to choose the safest next step.

Frequently Asked Questions about project-memory

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I recall prior codebase decisions before starting a new feature?

To recall prior codebase decisions, search the project memory index using tag overlap and grep fallback to retrieve bounded, evidence-aware context of past patterns and incidents.

What is evidence-gated project memory retrieval for software engineering?

Evidence-gated project memory retrieval is a workflow that blocks action when required memory artifacts or runtime proofs are missing, ensuring safe development without guessing.

Does project memory search work when the index is missing or incomplete?

Project memory search falls back to keyword grep when tag overlap fails, but it returns BLOCKED if required memory artifacts or expected inputs are entirely missing.

How to check staleness of past incident learnings before a refactor?

To check staleness of incident learnings, the memory recall process applies confidence and staleness checks to matched entries, emitting actionable recommendations with file path citations.