Gives AI coding agents a structural memory of your codebase, mapping API endpoints, symbol relationships, change impact, and pull request reviews through a local code graph. Includes skills for codebase exploration and co-change analysis.
Claude Code
> /plugin marketplace add syncable-dev/memtrace-public> /plugin install memtrace-skills@memtrace
Codex
$ npx skills add syncable-dev/memtrace-public -a codex
Cursor
$ npx skills add syncable-dev/memtrace-public -a cursor
Gemini CLI
$ npx skills add syncable-dev/memtrace-public -a gemini-cli
GitHub Copilot
$ copilot plugin marketplace add syncable-dev/memtrace-public$ copilot plugin install memtrace-skills@memtrace
OpenCode
$ npx skills add syncable-dev/memtrace-public -a opencode
Windsurf
$ npx skills add syncable-dev/memtrace-public -a windsurf
Official Memtrace by Syncable. Start at memtrace.io/docs. The official package is memtrace on npm, and the public repository is syncable-dev/memtrace-public. Other projects with the same name are unrelated. Compare memtrace --version with the Stable and Nightly release history.
For installation support, open an issue in the official public repository with your operating system, Memtrace version and exact error. Remove credentials and private source from shared logs. The Syncable team maintains this package and its support channels.
๐ Docs ย ยทย โญ Star us ย ยทย memtrace.io ย ยทย npm ย ยทย Discord
Memtrace turns your codebase into a live knowledge graph that AI coding agents can query in milliseconds โ every function, class, call edge, and version, across every session, without re-reading files or breaking things they can't see.
Get your fleet on shared structural memory in under 90 seconds.
Structural ยท zero LLM calls ย ยทย Bi-temporal ยท time-travel queries ย ยทย Replay-aware ยท zero blind refactors
Memtrace runs as a DeepSeek Harness plugin. Install Harness first (npm install -g @deepseek-ai/dsh โ that is the dsh command), then add Memtrace:
npx -y @deepseek-ai/dsh plugin --profile web add github:syncable-dev/dsh-plugin-memtrace
Then ask the agent to index the workspace and pull blast radius, evolution, or an architecture briefing. Details: syncable-dev/dsh-plugin-memtrace.
Three things, every release.
๐งญ ย Run a fleet of coding agents on the same repo without merge hell. Each agent reads the same call graph, sees the same blast radius, inherits the same temporal history. No collisions. No stale context.
๐ ย Replay any refactor with full causal awareness. Agents see exactly what depends on what, and what changed when. No more "I refactored a function and 14 tests broke that nobody saw."
โก ย Index a 50k-file repo in under 90 seconds. Rust + Tree-sitter, $0 in API costs, 20+ languages plus framework-aware scanners (Vapor, Lapis, Kong, GitHub Actions, Terraform, RLS policies, โฆ), fully local. Your code never leaves your machine.
๐ ย LeanCTX Native โ compressed reads, smart trees, and a value ledger.
Four new compression modes on get_source_window, single-call directory maps, real-time token-savings dashboard, and an opt-in adaptive learner that beats the static table by ~14%. Full breakdown: docs/leanctx-native.md. Available in v0.3.57+.
| Operation | Memtrace | Best alternative | ฮ |
|---|---|---|---|
| Index 1,500 files | 1.5s ยท $0 | Mem0: 31 min ยท $10โ50 | ~1,200ร faster |
| Exact symbol query (acc@1, lat) | 96.6% ยท 0.07 ms | GitNexus: 97.0% ยท 8.95 ms | 128ร lower latency |
| Graph callers recall (Django) | 81.6% | GitNexus: 5.3% | 15.4ร |
| Incremental re-index p95 | 42.5 ms | CodeGrapher: 613.7 ms | 14.4ร |
| Hybrid acc@1 (Django, 3K cases) | 73.9% | GitNexus: 38.6% | 1.91ร |
| PR code-review F1 (50 PRs) | 0.7268 | Cubic v2: 0.6077 | +19.60% |
| RSS / process | 26 MB | ChromaDB: 1,060 MB | 41ร tighter |
| Languages | 16+ (Tree-sitter) | varies | โ |
Reproducible benchmark suite: benchmarks/. Same machine, same corpora, same adapter contract. Ground truth from Python's ast and pyright LSP โ never from any tool's own index. No system gets a home-field advantage in the dataset.
Detailed breakdowns: BENCHMARKS-v0.3.22.md ยท BENCHMARKS-v0.3.29.md ยท Code reviewer benchmark
Memtrace is in private beta. We're rolling out access in batches to keep the feedback loop tight โ every cohort lands in a Discord channel where we ship fixes from real bug reports inside a week.
โ Join the waitlist at memtrace.io.
Already have access? npm install -g memtrace and you're indexing in 90 seconds. Full setup below.
๐ Privacy. Memtrace runs entirely on your machine. Source code never leaves it. The only network traffic is license validation, aggregate node/edge counts, and opt-out crash telemetry โ no source, no file paths, no symbol names. Full breakdown: PRIVACY.md, TELEMETRY.md. Disable telemetry with
MEMTRACE_TELEMETRY=off.
Good code-intelligence tools already exist. GitNexus and CodeGrapherContext build AST-based graphs that work for "what's in my repo right now."
Memtrace is a bi-temporal episodic structural knowledge graph. It builds on the same AST foundation and adds two dimensions:
On top of that, the structural layer is comprehensive:
| Symbols are nodes | functions, classes, interfaces, types, endpoints |
| Relationships are edges | CALLS, IMPLEMENTS, IMPORTS, EXPORTS, CONTAINS |
| Community detection | Louvain algorithm identifies architectural modules automatically |
| Hybrid retrieval | Tantivy BM25 + vector embeddings + Reciprocal Rank Fusion + cross-encoder rerank |
| Rust-native | compiled binary, no Python/JS runtime overhead, sub-8 ms p95 query latency |
The agent doesn't just search your code. It remembers it.
Mem0 and Graphiti are strong conversational memory engines designed for tracking entity knowledge (e.g. User -> Likes -> Apples). They excel at that. For code intelligence specifically, the tradeoff is that they rely on LLM inference to build their graphs โ which adds cost and time when processing thousands of source files.
Graphiti processes data through add_episode(), which triggers multiple LLM calls per episode โ entity extraction, relationship resolution, deduplication. At ~50 episodes/minute (source), ingesting 1,500 code files takes 1โ2 hours.
Mem0 processes data through client.add(), which queues async LLM extraction and conflict resolution per memory item (source). Bulk ingestion with infer=True (default) means every file passes through an LLM pipeline. Throughput is bounded by your LLM provider's rate limits.
Both accumulate $10โ50+ in API costs for large codebases because every relationship is inferred rather than parsed.
Memtrace takes a different approach: it indexes 1,500 files in 1.2โ1.8 seconds for $0.00 โ no LLM calls, no API costs, no rate limits. Native Tree-sitter AST parsers resolve deterministic symbol references (CALLS, IMPLEMENTS, IMPORTS) locally. The tradeoff is that Memtrace is purpose-built for code โ it doesn't handle conversational entity memory the way Mem0 and Graphiti do.
Memtrace exposes a full structural toolkit via the Model Context Protocol.
Search & Discovery
find_code โ hybrid BM25 + semantic + RRFfind_symbol โ exact / fuzzy with LevenshteinRelationships
analyze_relationships โ callers, callees, hierarchy, importsget_symbol_context โ 360ยฐ view in one callImpact Analysis
get_impact โ blast radius with risk ratingdetect_changes โ diff-to-symbols scope mappingCode Quality
find_dead_code โ zero-caller detectionfind_most_complex_functions โ complexity hotspotscalculate_cyclomatic_complexityget_repository_statsTemporal Analysis
get_evolution โ 6 scoring modesget_timeline โ full version historydetect_changes โ diff-based scopeGraph Algorithms
find_bridge_symbols โ betweenness centralityfind_central_symbols โ PageRank / degreelist_communities โ Louvain moduleslist_processes / get_process_flowAPI Topology
get_api_topology โ cross-repo HTTP graphfind_api_endpointsfind_api_callsIndexing & Watch
index_directory โ parse, resolve, embedwatch_directory โ live incrementalexecute_cypher โ direct graph queriesMemtrace ships skills/guidance that teach agents how to use the graph. They fire automatically based on what you ask โ no prompt engineering required.
| Skill | You sayโฆ |
|---|---|
memtrace-search | "find this function", "where is X defined" |
memtrace-relationships | "who calls this", "show class hierarchy" |
memtrace-evolution | "what changed this week", "how did this evolve" |
Stars on GitHub
A development methodology for coding agents that makes the agent brainstorm before building, follow plans, run parallel work and finish branches in a disciplined way. You get fewer skipped steps and more consistent results from your coding assistant.
@obraHelps you take a change from a rough idea or bug report through to a tested, reviewed result, with each step handled by a focused skill. Its skills cover pinning down domain terms, designing module interfaces, diagnosing hard bugs, and reviewing changes against your repo's coding standards.
@mattpocockGives your coding agent a planning, testing and review routine to follow on every task. It plans before building, checks changes with tests, has a fresh reviewer look for regressions, and saves repeated wins as reusable skills.
@affaan-mDeveloper tools for working on the React codebase itself. Provides skills for extracting error messages, managing and debugging feature flags across release channels, running Flow type checks, and fixing lint and formatting issues before committing.
@facebookInstall it with the command on its page. It includes memtrace-api-topology, memtrace-change-impact-analysis, memtrace-cochange. Its skills do not fire on their own yet. Free and open source.
Is this plugin yours?
Claim it with GitHubSubmit a pluginPromote it