conport-memory_skill

This skill maintains persistent project memory across sessions by loading context, logging decisions, and tracking progress to prevent AI amnesia.

2

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill nicobailon/conport-mcporter-skills --skill conport-memory

  • SKILL.md6.9 KB

Overview

This skill connects your assistant to ConPort (Context Portal) as a persistent project memory via the MCPorter CLI. It stores and retrieves product and active context, decisions, progress, patterns, and custom project data in a SQLite-backed knowledge base. Use it to prevent AI amnesia across sessions and to keep a searchable, linked history of project choices and outcomes.

How this skill works

The skill calls ConPort tools through MCPorter commands to load and update context and to log events. At session start it can load product_context, active_context, in-progress tasks, and recent decisions. During work it logs decisions, progress, system patterns, and custom key-value data. It also supports linking items to build a lightweight knowledge graph and exporting the entire ConPort database to Markdown.

When to use it

  • At session start to load product_context and active_context and resume work
  • When making an architectural or design choice to log decision, rationale, and tags
  • When completing or updating tasks to log_progress and track status
  • When discovering a reusable solution to log_system_pattern for future reuse
  • At session end to update_active_context with current state and next steps

Best practices

  • Proactively load product and active context at the beginning of each session
  • Log concise decisions with rationale and tags to make them discoverable later
  • Use status values (e.g., in_progress, completed) when logging progress for clear filtering
  • Record system patterns with names and tags to encourage reuse across teams
  • Link decisions, progress, and patterns to form relationships that clarify implementation traceability

Example use cases

  • Resume a sprint by loading active_context and in-progress tasks to avoid repeating orientation
  • Record why a database or auth approach was chosen, with tags for easy search later
  • Mark a task complete and automatically link it to the decision that drove its implementation
  • Capture a discovered repository or API pattern so future engineers can apply it
  • Export the entire project context to Markdown for handoff or audit purposes

FAQ

Run get_product_context and get_active_context via the MCPorter base command to fetch goals and the current working focus.

Can I search past decisions for a keyword?

Yes. Use search_decisions_fts with a query string to full-text search across logged decisions.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational