[Trend Scout] sara-star-quant/presence

[github-actions]

🔭 Trend Scout: sara-star-quant/presence

Auto-generated by trend-scout.py — review and edit as needed.

📌 What problem it solves

Per-repo memory, outcome telemetry, and a calibrated-confidence gate for Claude Code, with MCP and AGENTS.md projections so other AI coding tools can read its context. Notes survive sessions; success claims need test evidence; your reverts are remembered. Local-only, stdlib runtime.

📅 Timeline

Field	Value
Created	2026-04-25
Last pushed	2026-06-03
Stars	2
Forks	0
Open issues	4
License	MIT
Language	Python
Topics	agent-memory, agentic-coding, agents-md, ai-pair-programming, ai-tools, anthropic, calibrated-confidence, claude, claude-code, claude-code-plugin, coding-assistant, compliance-redaction, developer-productivity, hooks, llm-tooling, mcp, mcp-server, prompt-engineering, python3, session-continuity

✅ Strengths

• Well-tagged: agent-memory, agentic-coding, agents-md, ai-pair-programming, ai-tools, anthropic
• Primary language: Python
• Actively maintained (pushed within 30 days)

⚠️ Weaknesses / Risks

• Low star count (2) — limited community validation

💡 What this repo can learn

• Claude Code session patterns: this repo's Claude Code integration approach could improve claude-adapter.py's JSONL parsing — e.g., handling new session event types or extracting richer metadata from Claude Code tool-use blocks
• CLI verb patterns: a clear add/search/update/delete verb model (like memory-tool add / search / dream) could streamline the UX of query-session.py and learn.py, making them easier to invoke from hooks or scripts
• Git hook / workflow patterns: hook design from this repo could strengthen the hooks/ enforcement chain (e.g., auto-briefing, commit guards, learn reminders)
• MCP tool-server surface: the MCP server interface here could expose query-session.py and briefing.py as directly callable MCP tools — e.g., a Copilot agent could invoke the briefing lookup in-process rather than shelling out to a subprocess, reducing latency in hook-driven workflows

README excerpt

# presence

[![CI](https://github.com/sara-star-quant/presence/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/sara-star-quant/presence/actions/workflows/ci.yml)
[![Latest release](https://img.shields.io/github/v/release/sara-star-quant/presence?sort=semver&cacheSeconds=60)](https://github.com/sara-star-quant/presence/releases)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
[![Python 3.12+](https://img.shields.io/badge/python-3.12%20%7C%203.13%20%7C%203.14-blue)](.github/workflows/ci.yml)
[![Stdlib only](https://img.shields.io/badge/runtime-stdlib--only-success)](pyproject.toml)
[![Local only](https://img.shields.io/badge/state-local--only-success)](docs/security.md)

**Every Claude Code session starts cold. presence makes the next one start where the last one left off.**

A Claude Code plugin with read-only projections (MCP server, AGENTS.md adapter) so MCP-aware clients (Cursor, Claude Desktop, Continue) and AGENTS.md-aware tools (Codex, Gemini CLI, Windsurf, GitHub Copilot) can also read its accumulated context. Turns every session into part of a continuum.

`presence` adds four things to Claude Code, globally, with one install and zero per-project setup:

1. **Living project model.** Claude builds and reuses notes about each repo it touches. No more re-deriving the same architecture every session.
2. **Outcome telemetry.** Tracks what Claude committed, then watches for reverts, amends, and PR closes. Future sessions see

*(truncated)*

---

Scouted on 2026-06-03 · View on GitHub

[m13v]

presence rebuilds the next session from notes. that's recall, not continuity. the live context window is exactly what cold-start drops.