Skills2026-08-09 · source-backed

Index the agent session logs already on your disk instead of building a memory system.

Issue

2026-08-09

Confidence

source-backed

Sources

Redaction

redacted

Story

deja-vu ships as an MCP server that searches your existing Claude Code, Codex, and other transcripts, so any MCP-speaking harness can query months of prior work with no re-ingestion. The corpus already exists and cost nothing to produce. Check ~/.claude/projects/ before you build a retain/recall pipeline you don't need.

Codex competes with Claude Code / Shared entities / Same source domain / Shared topic / Earlier coverage
Y Combinator open-sources QM, the harness running its company
Linked by a graph relationship (Codex competes with Claude Code); both cover Claude Code, Codex, MCP; reported by the same outlet (github.com).
Agent memory consolidated into a cross-harness layer nobody owns.
Linked by a graph relationship (Codex competes with Claude Code); both cover Claude Code, Codex, MCP; reported by the same outlet (github.com).
Skills supports Codex / Shared entities / Same source domain / Shared topic / Earlier coverage / Tension
The Claude Code Field Guide Hit 53,400 Stars. Here's Why That Matters.
Linked by a graph relationship (Skills supports Codex); both cover Claude Code, Codex, MCP; reported by the same outlet (github.com).
Codex competes with Claude Code / Shared entities / Same source domain / Shared topic / Earlier coverage
Sift collapses hundreds of MCP tool definitions down to two exposed tools.
Linked by a graph relationship (Codex competes with Claude Code); both cover Claude Code, Codex, MCP; reported by the same outlet (github.com).
context-mode sandboxes tool output entirely for a claimed 98% reduction across 17 platforms.
Linked by a graph relationship (Codex competes with Claude Code); both cover Claude Code, Codex, MCP; reported by the same outlet (github.com).
Personal Model ships an evidence-linked HUMAN.md built from captured Mac activity.
Linked by a graph relationship (Codex competes with Claude Code); both cover Claude Code, Codex, MCP; reported by the same outlet (github.com).
Terminal dashboard manages four AI coding agents
Linked by a graph relationship (Codex competes with Claude Code); both cover Claude Code, Codex, MCP; reported by the same outlet (github.com).
ai-memory compiles a git-versioned wiki at session end instead of retrieving over raw logs.
Linked by a graph relationship (Codex competes with Claude Code); both cover Claude Code, Codex, MCP; reported by the same outlet (github.com).

Source trail

deja-vu

Entities

Check Claude Code Codex Index MCP

Provenance

Canonical issue: Ramsay Research Agent — August 9, 2026
AI generated: no
Story unit: 2026-08-09-index-the-agent-session-logs-already-on-your-disk-instead-of-building-a-memory-system
Labels: source-backed, canonical briefing excerpt

Fetching from the wire…

Index the agent session logs already on your disk instead of building a memory system.

Story

Related stories

Y Combinator open-sources QM, the harness running its company

Agent memory consolidated into a cross-harness layer nobody owns.

The Claude Code Field Guide Hit 53,400 Stars. Here's Why That Matters.

Sift collapses hundreds of MCP tool definitions down to two exposed tools.

context-mode sandboxes tool output entirely for a claimed 98% reduction across 17 platforms.

Personal Model ships an evidence-linked HUMAN.md built from captured Mac activity.

Terminal dashboard manages four AI coding agents

ai-memory compiles a git-versioned wiki at session end instead of retrieving over raw logs.

Source trail

Entities

Provenance

Index the agent session logs already on your disk instead of building a memory system.

Story

Related stories

Y Combinator open-sources QM, the harness running its company

Agent memory consolidated into a cross-harness layer nobody owns.

The Claude Code Field Guide Hit 53,400 Stars. Here's Why That Matters.

Sift collapses hundreds of MCP tool definitions down to two exposed tools.

context-mode sandboxes tool output entirely for a claimed 98% reduction across 17 platforms.

Personal Model ships an evidence-linked HUMAN.md built from captured Mac activity.

Terminal dashboard manages four AI coding agents

ai-memory compiles a git-versioned wiki at session end instead of retrieving over raw logs.

Source trail

Entities

Provenance