agent-knowledge

Cross-session memory and recall for AI coding assistants -- works with Claude Code, Cursor, OpenCode, Cline, Continue.dev, and Aider out of the box. Git-synced knowledge base, hybrid semantic+TF-IDF search, auto-distillation with secrets scrubbing.

Benchmark: R@5 = 97.2% (sparse) / 98.8% (hybrid) on longmemeval_s and 86.0% (sparse) / 88.4% (hybrid) on the harder longmemeval_m split — the public LongMemEval academic benchmark (Wu et al. 2024, ICLR 2025), full 500 questions per split, no LLM, no API key, runs entirely offline. +8.6pp to +13.2pp R@5 over the paper's official flat-bm25 baseline in apples-to-apples reproduction. Full per-category table, reproduction instructions, and paper-comparison details in bench/README.md.

Why

AI coding sessions are ephemeral. When a session ends, everything it learned -- architecture decisions, debugging insights, project context -- is gone. The next session starts from scratch.

agent-knowledge solves this with two complementary systems:

1. Knowledge Base -- a git-synced markdown vault of structured entries (decisions, workflows, project context) that persists across sessions and machines.
2. Session Search -- TF-IDF ranked full-text search across session transcripts from all your coding tools, so agents can recall what happened before -- regardless of which tool was used.

Supported Tools

Sessions from all major AI coding assistants are auto-discovered -- if a tool is installed, its sessions appear automatically.

Tool	Format	Auto-detected path
Claude Code	JSONL	~/.claude/projects/
Cursor	JSONL	~/.cursor/projects/*/agent-transcripts/
Codex CLI	JSONL	~/.codex/projects/
Aider	Markdown/JSONL	.aider.chat.history.md / .aider.llm.history in project dirs
Continue.dev	JSON	~/.continue/projects/
Cline	JSON	VS Code globalStorage saoudrizwan.claude-dev/tasks/
OpenCode	SQLite	~/.local/share/opencode/opencode.db (or $OPENCODE_DATA_DIR)

No configuration needed. Additional session roots can be added via the AGENT_KNOWLEDGE_EXTRA_SESSION_ROOTS env var (comma-separated paths).

Features

• Host-agnostic session search -- unified search across every major AI coding assistant (Claude Code, Cursor, Codex CLI, Aider, Continue.dev, Cline, OpenCode). No host name is baked into configuration — the adapter registry probes installed host roots at startup.
• Hybrid search -- semantic vector similarity blended with TF-IDF keyword ranking
• Git-synced knowledge base -- markdown vault with YAML frontmatter, auto commit and push on writes
• Automatic staleness detection -- knowledge_analyze(action: "stale_by_code_activity") cross-references file paths mentioned in each entry body against filesModified in recent session summaries. Pairs with a symbol-presence precision layer: identifiers the entry quotes (inline backticks + fenced blocks) are checked in the touched file; if they still exist, confidence downweights ×0.3. Entries with evergreen: true are exempt.
• Search-gap tracking -- knowledge_analyze(action: "search_gaps") surfaces zero-result queries over the last since_days, grouped by token-Jaccard similarity. The clearest signal for "what entries should I write next?".
• Section-priority context packer -- knowledge(action: "wakeup") assembles a multi-section bundle (identity → active_tasks → recent_decisions → known_gotchas → last_session_summary → top_weighted → semantic_fallback) within a token budget (default 800, override via token_budget or AGENT_KNOWLEDGE_WAKEUP_BUDGET). Unused section budget redistributes to later sections.
• Scored + gated promoter -- session insights promoted via a 6-signal weighted scorer with three independent gates (minScore, minRecallCount, minUniqueQueries). Runs automatically in background, on demand via knowledge_admin(action: "promote"), or benchable offline via npm run bench:promote. Emits an auditable .dreams/YYYY-MM-DD.md diary every run.
• Pluggable adapter system -- add support for new tools by implementing the SessionAdapter interface
• Embeddings -- local (Hugging Face), OpenAI, Claude/Voyage, or Gemini providers
• Fuzzy matching -- typo-tolerant search using Levenshtein distance
• 6 search scopes -- errors, plans, configs, tools, files, decisions
• 6 MCP tools -- consolidated action-based interface (knowledge, knowledge_search, knowledge_session, knowledge_graph, knowledge_analyze, knowledge_admin)
• Evergreen entries -- evergreen: true in frontmatter exempts an entry from decay in ranking AND makes it append-only under promotion. Dashboard renders a push-pin badge on these cards.
• Author attribution -- optional author: <string> frontmatter surfaces as a muted chip on each card.
• Code graph resolution -- calls, imports, inherits edge types for code structure; directed BFS traversal (outbound/inbound/both); bulk_link for efficient ingestion; unlink_by_origin for clearing stale code edges before re-ingest; code: prefixed node IDs distinguish code from knowledge
• Temporal knowledge graph -- edges support valid_from / valid_to validity windows; as_of queries return point-in-time snapshots; invalidate action marks facts as ended without deleting them
• Hybrid scoring boosts -- proper-noun and temporal-proximity boosts on top of TF-IDF + semantic blend, capped at +66.7%, short-circuit when no signals are present
• Category as boost (not filter) -- opt into category_mode: "boost" so a wrong category guess down-ranks instead of discarding the right answer
• Verbatim session indexing -- per-message chunks (≥30 chars) embedded into the vector store so raw conversation is retrievable; toggle with AGENT_KNOWLEDGE_INDEX_VERBATIM=false
• Configurable git URL -- knowledge_admin(action: "config") for runtime setup, persisted at XDG/AppData location
• Cross-machine persistence -- knowledge syncs via git, sessions read from local storage of each tool
• Real-time dashboard -- browse, search, and manage at localhost:3423
• Secrets scrubbing -- API keys, tokens, passwords, private keys automatically redacted before git push
• Knowledge graph -- relationship edges between entries (related_to, supersedes, depends_on, contradicts, specializes, part_of, alternative_to, builds_on) with BFS traversal
• Confidence/decay scoring -- entries scored by access frequency and recency; auto-promotion from candidate to established to proven
• Memory consolidation -- TF-IDF duplicate detection on write (warns of similar entries) plus knowledge_analyze(action: "consolidate") for batch dedup scanning
• Reflection cycle -- knowledge_analyze(action: "reflect") surfaces unconnected entries and generates structured prompts for the agent to identify new graph connections
• Auto-linking on write -- new entries automatically linked to top-3 similar existing entries when cosine similarity > 0.7
• Confidence metadata — entries tagged extracted (user-written) or inferred (auto-distilled, 0.85× search rank multiplier); confidence_score field carries the model's certainty 0-1
• Knowledge analysis — knowledge_analyze actions god_nodes (most-connected entries), bridges (cross-category connectors), gaps (isolated entries)
• Knowledge brief — knowledge_analyze(action: "brief") returns a cached ~200 token summary (core concepts, active projects, recent decisions, stale and gap counts) for session-start orientation
• Edge provenance — graph edges track origin (manual, auto-link, distill, reflect) so analysis can distinguish user judgment from automated heuristics
• Deterministic pre-extraction in distillation — session summaries now include git commits, error patterns, URLs accessed, and packages changed extracted via regex from bash/tool output (no LLM cost)
• Freshness metadata on every search hit — every knowledge result carries freshness: { body_age_days, last_accessed, access_count, verified_at, verification_age_days, evergreen }. Agent reads the trust signal and decides; we impose no policy demotion.
• Per-category decay windows — the "Unused" filter and bytype chart honor per-category thresholds (projects 180d, people 365d, decisions 90d, workflows 60d, notes 30d) so identity-shaped content doesn't look stale just because it isn't re-read weekly.
• Lifecycle hooks — SessionStart auto-wakeup + ingest-freshness check, UserPromptSubmit first-prompt targeted injection, PreCompact memory-flush nudge + distill, SessionEnd distill. Six hook scripts total, all fail-open, each toggleable via an AGENT_KNOWLEDGE_* env var. See docs/HOOKS.md.
• Replaces host auto-memory — on hosts with a per-session memory system (Claude Code's ~/.claude/projects/*/memory/, similar in other IDEs), route durable user facts and feedback to agent-knowledge instead. Auto-memory is machine-local and invisible to other machines; agent-knowledge is git-synced, cross-machine, searchable, and surfaces in wakeup. See the Claude Code integration note in docs/USER-MANUAL.md.

Codebase Ingestion

The knowledge-ingest skill populates or updates the knowledge base from a codebase directory. It uses tree-sitter for zero-token structural extraction (classes, functions, imports, call graphs, rationale comments), then clusters files into subsystems and creates knowledge entries + graph edges via existing MCP tools. Subsequent runs are incremental — only changed files are reprocessed.

/knowledge-ingest ./my-project

Uses the Agent Skills standard — works with Claude Code, OpenCode, Cursor, Codex CLI, and Gemini CLI. See Ingestion Guide for details.

Supported languages: TypeScript, JavaScript, Python, Go, Rust, Java, C, C++.

Quick Start

Install from npm

npm install -g agent-knowledge

Or clone from source

git clone https://github.com/keshrath/agent-knowledge.git
cd agent-knowledge
npm install && npm run build

Option 1: MCP server (for AI agents)

Add to your MCP client config (Claude Code, Cline, etc.):

{
  "mcpServers": {
    "agent-knowledge": {
      "command": "npx",
      "args": ["agent-knowledge"]
    }
  }
}

The dashboard auto-starts at http://localhost:3423 on the first MCP connection.

See Setup Guide for client-specific instructions (Claude Code, Cursor, Windsurf, OpenCode).

Option 2: Standalone server (for REST/WebSocket clients)

node dist/server.js --port 3423

MCP Tools (6)

Knowledge Base

Tool	Action	Description	Parameters
knowledge	list	List entries by category and/or tag	category?, tag?
	read	Read a specific entry	path (required)
	write	Create/update entry (auto git sync)	category, filename, content (all required)
	delete	Delete an entry (auto git sync)	path (required)
	sync	Manual git pull + push	--
	wakeup	Return L0 identity + L1 top-weighted entries (token-budgeted)	token_budget?, category?

Search

Tool	Description	Parameters
knowledge_search	General hybrid TF-IDF + semantic (no scope)	query, project?, role?, max_results?, ranked?, semantic?, category?, category_mode?, mmr?, mmr_lambda?, explain?
	Scoped session-only recall (when scope set)	query, scope, project?, max_results?

Response shape: {mode: "general" | "scoped", sessions, knowledge}. Scoped mode returns knowledge: [] by design.

Scopes: errors, plans, configs, tools, files, decisions, all.

Search knobs:

• mmr: true applies Maximal Marginal Relevance re-ranking (kills near-duplicate clusters in top-K). mmr_lambda 0-1, default 0.7.
• category_mode: "boost" (default) gives matching-category entries a 1.25× score multiplier instead of dropping non-matches. Pass "filter" for hard-filter behavior.
• explain: true attaches score_components: {bm25, decay, maturity, confidence, category_boost, mmr_penalty} to every knowledge hit.

Sessions

Tool	Action	Description	Parameters
knowledge_session	list	List sessions with metadata	project?
	get	Retrieve full session conversation	session_id, project?, include_tools?, tail?
	summary	Session summary (topics, tools, files)	session_id, project?

Knowledge Graph

Tool	Action	Description	Parameters
knowledge_graph	link	Create/update edge between entries	source, target, rel_type, strength?
	unlink	Remove edges between entries	source, target, rel_type?
	invalidate	Mark edges as expired (set valid_to)	source, target, rel_type?, valid_to?
	list	List edges	entry?, rel_type?, as_of?
	traverse	Directed BFS traversal from an entry	entry, depth?, direction?, rel_type?, as_of?
	bulk_link	Batch-create edges (code graph ingestion)	edges (array of {source, target, rel_type, strength?, origin?})
	unlink_by_origin	Delete all edges by origin	origin

Knowledge types: related_to, supersedes, depends_on, contradicts, specializes, part_of, alternative_to, builds_on Code structure types: calls, imports, inherits

Traverse directions: outbound (source→target), inbound (target→source), both (default, undirected)

Analysis

Tool	Action	Description	Parameters
knowledge_analyze	consolidate	Find near-duplicate entries	category?, threshold?
	reflect	Find unconnected entries for linking	category?, max_entries?
	god_nodes	Most-connected entries (degree centrality)	top_n?
	bridges	Cross-category connectors (betweenness)	top_n?
	gaps	Isolated entries (0-1 edges) by maturity	max_entries?
	brief	Cached ~200 token knowledge base summary	--

Admin

Tool	Action	Description	Parameters
knowledge_admin	status	Vector store statistics	--
	config	View or update configuration	git_url?, memory_dir?, auto_distill?
	rebuild_embeddings	Re-embed all knowledge entries (useful on provider switch)	--
	prune_orphans	Delete embeddings for sessions no longer on disk	vacuum?, force_vacuum?
	vacuum	Reclaim free pages in the vector store	--
	promote	Scored + gated promoter	promote_mode? (apply|explain), min_score?, min_recall_count?, min_unique_queries?

Scored promoter

Every project-level candidate is scored on six signals (relevance 0.30, frequency 0.24, query-diversity 0.15, recency 0.15, consolidation 0.10, conceptual-richness 0.06) and gated on minScore ≥ 0.5, minRecallCount ≥ 2, minUniqueQueries ≥ 2. All three gates must pass. Background auto-promotion is controlled by the same auto_distill config flag; invoke on demand with knowledge_admin(action: "promote").

• promote_mode: "explain" (default) — score + gate candidates, write diary, DO NOT touch the KB.
• promote_mode: "apply" — promote candidates that pass, write diary, git-commit.
• Every run drops ~/agent-knowledge/.dreams/YYYY-MM-DD.md with per-candidate signal breakdowns and gate outcomes. The .-prefixed dir is git-tracked but excluded from list/search.
• Grounded rehydration: a candidate is skipped if its source session file no longer exists on disk (prevents promoting deleted content).
• Entries with evergreen: true frontmatter are never overwritten by promotion — activity is appended.

Write-bench harness: npm run bench:promote — offline replay with auto-labeling by "referenced in later sessions". Compares gated promoter to a naive "ship all" baseline, reports precision / recall / F1. Use it to gate signal-weight or threshold changes before rolling them out.

REST API

Method	Endpoint	Description
GET	/api/knowledge	List knowledge entries
GET	/api/knowledge/search?q=	Search knowledge base
GET	/api/knowledge/:path	Read a specific entry
GET	/api/knowledge/god-nodes?top_n=	Most-connected entries
GET	/api/knowledge/bridges?top_n=	Cross-category connectors
GET	/api/knowledge/gaps?max_entries=	Isolated entries
GET	/api/knowledge/brief	Knowledge base brief
GET	/api/sessions	List sessions
GET	/api/sessions/search?q=&role=&ranked=	Search sessions (TF-IDF)
GET	/api/sessions/recall?scope=&q=	Scoped recall
GET	/api/sessions/:id	Read a session
GET	/api/sessions/:id/summary	Session summary
POST	/api/knowledge	Write entry (HTTP clients)
GET	/health	Health check

Architecture

graph LR
    subgraph Storage
        KB[(Knowledge Base<br/>~/agent-knowledge<br/>Git Repository)]
    end

    subgraph Session Sources
        CC[(Claude Code<br/>JSONL)]
        CU[(Cursor<br/>JSONL)]
        OC[(OpenCode<br/>SQLite)]
        CL[(Cline<br/>JSON)]
        CD[(Continue.dev<br/>JSON)]
        AI[(Aider<br/>MD / JSONL)]
    end

    subgraph agent-knowledge
        KM[Knowledge Module<br/>store / search / git]
        AD[Session Adapters<br/>auto-discovery]
        SE[Search Engine<br/>TF-IDF + Fuzzy]
        DS[Dashboard<br/>:3423]
        MCP[MCP Server<br/>stdio]
    end

    subgraph Clients
        AG[Agent Sessions]
        WB[Web Browser]
    end

    KB <-->|git pull/push| KM
    CC --> AD
    CU --> AD
    OC --> AD
    CL --> AD
    CD --> AD
    AI --> AD
    AD --> SE
    KM --> MCP
    SE --> MCP
    KM --> DS
    SE --> DS
    MCP --> AG
    DS --> WB

Knowledge Graph

Entries and code symbols can be connected via typed, weighted edges stored in a dedicated edges SQLite table. Eleven relationship types are supported — 8 for knowledge edges and 3 for code structure:

Knowledge: related_to, supersedes, depends_on, contradicts, specializes, part_of, alternative_to, builds_on Code structure: calls, imports, inherits