Score8.6
Popularity95.0
Risklow
TierGold
Score breakdown
Usefulness9.0
Novelty9.0
Momentum7.0
Maturity8.7
Open-source/build8.4
Evidence8.0
Workflow potential9.7
Setup ease8.8
Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.
Why it matters
Useful for **AI coding agent users (Claude Code / Codex / Cursor / Copilot / Aider)** — projectmem is the MIT local-first memory + judgment layer that sits above your AI tools, captures every failed attempt, decision, and gotcha, and injects that experience back into future AI sessions. Useful for **anyone tired of watching their AI repeat the same failed approach in a new session** — `pjm prechec
Who should use it
**AI coding agent users (Claude Code / Codex / Cursor / Copilot / Aider)** — projectmem is the MIT local-first memory + judgment layer that sits above your AI tools, captures every failed attempt, decision, and gotcha, and injects that experience back into future AI sessions**Anyone tired of watching their AI repeat the same failed approach in a new session** — `pjm precheck` warns *before* commit if you're about to repeat a dead-end approach ('What already failed here: ✗ tried CSS contain:layout'), modify a high-churn file, or touch an unresolved issue. No other AI tool does this — it requires the memory layer underneath**Teams that need memory to survive across projects** — Cross-Project Memory means library gotchas, decisions, and patterns live in `~/.projectmem/global/` and auto-inherit into every new project that matches your stack**Teams using MCP-capable AI agents** — Native MCP server with 14 tools (verified end-to-end against Claude Desktop, Cursor, Antigravity, Codex)**Teams using non-MCP agents (Copilot, plain Claude, Aider)** — `pjm export --claude-md` compiles live decisions, gotchas, and a 'Do NOT retry — these already failed' list into a marked block in CLAUDE.md / `.cursorrules`**Teams that need a plain-text greppable store (not a SaaS)** — events.jsonl is for tools; AI reads small distilled files. Tools distill the raw log into compact AI-readable summaries**Teams that need provable ROI metrics** — `pjm score` outputs a letter grade (A+ → F) backed by concrete numbers (debugging hours saved, tokens prevented, dollars protected). CI-friendly JSON output and shields.io badge**Token-efficiency-conscious users** — MCP mode is ~800–1,500 tokens/session (vs. baseline 5,000–20,000+ for AI re-reading source files)**Teams that want their memory audit trail** — every CLI write makes one git commit, the history is the audit trail, and `pjm visualize` opens a four-tab D3.js dashboard (Story Map / ROI Dashboard / Project Map / Timeline)**MIT-licensed local-first memory** — plain MIT, no per-file carve-outs, no SaaS-embedding caveat, no commercial-use threshold, 100% local, no telemetry, no accountsEvaluation: `pip install projectmem && cd your-project && pjm init`. For testing the pre-commit warning: try to commit a change to a file that has known failed approaches in the memory — `pjm precheck` blocks the commit with the dead-end list. For testing the stale-memory detection: edit a file that has a memory citing it, then ask `pjm brief` — the memory will be flagged as 'predates 7 commits to auth.py — confirm or supersede'. For testing cross-project memory: `pjm decision 'my new way' --supersedes <id>` retires the old event cleanly while preserving it in the log. For MCP integration (Claude Desktop): add the projectmem server block to `claude_desktop_config.json` with `args=['-m', 'projectmem.mcp_server', '--root', '/absolute/path/to/project']`. Use the absolute path to `python` (run `which python`) — Claude Desktop subprocesses don't inherit shell PATH. Pass `--root`, not `cwd` (the field is silently ignored on Claude Desktop's Epitaxy/Cowork build). For MCP integration (Cursor / Codex / Antigravity): same pattern, drop into `<project>/.cursor/mcp.json` or `~/.codex/config.toml`. Set Codex reasoning effort to `medium` or higher so the session-start trio fires reliably. For the research backing: arXiv 2606.12329 'PROJECTMEM: A Local-First, Event-Sourced Memory and Judgment Layer for AI Coding Agents' — peer-readable design + 207-event dogfooding study. For the open-source status: plain MIT, riponcm / projectmem contributors, no per-file carve-out, no SaaS-embedding caveat, no commercial-use threshold. For the maturity caveat: at 95 stars and 3 forks, projectmem is a relatively new project — fine for early adoption, monitor releases for stability before depending on it as a long-term memory store
Who should skip it
Skip riponcm/projectmem unless the captured evidence suggests it solves a problem you are actively working on.
About this signal
riponcm/projectmem is tracked by RepoRadar as a mit local-first memory + judgmen in the riponcm/projectmem (MIT, riponcm / projectmem co section. It was first seen on 2026-06-26 and last updated on 2026-06-26. The current verdict is 'try now' with a Gold tier and easy setup difficulty. riponcm/projectmem leads on workflow potential (9.7) and practical usefulness (9.0); its lowest signal is momentum (7.0), so factor that in before investing setup time. This page summarizes the evidence RepoRadar has captured from captured source metadata. The score, tier, risk label, and verdict on this page are never influenced by sponsorship, ads, or tips — they reflect only the usefulness, popularity, novelty, momentum, maturity, and evidence signals described in the RepoRadar methodology.
How this item is evaluated
RepoRadar assigned riponcm/projectmem a composite score of 8.6 out of 10, placing it in the Gold tier. This score combines weighted sub-signals: usefulness (35%), novelty (18%), momentum (14%), maturity (10%), open-source/build quality (7%), evidence quality (6%), workflow potential (6%), and setup ease (4%). Popularity is tracked separately at 95.0 and never affects the composite score or tier. The risk label of 'low' reflects inherent user-impacting hazards, not generic novelty. Items with no risk flag may still require normal code review before production use.
Risk explanation
maturity caveat — at 95 stars and 3 forks; projectmem is a relatively new project; fine for early adoption; monitor releases for stability before depending on it as a long-term memory store.
Evidence links
Closest alternatives / related signals
projectmemriponcmlocal-first-memoryjudgment-layerai-coding-agentspre-commit-warningsstale-memory-detectionnever-delete-flag
