Continuity — Persistent Memory for AI Coding

Every AI session starts from zero. Your assistant forgets the database you chose last week, the pattern you rejected yesterday, and the constraint your team agreed on this morning.

Continuity fixes this. It gives AI tools structured, repo-local memory so they recover context instead of starting cold. Works via instruction files, CLI, and optionally MCP.

How It Works

1. Log decisions as you work — manually, via CLI, or auto-detected from git commits
2. AI recalls them across sessions via instruction files (CLAUDE.md, .cursorrules, etc.)
3. Decisions live in your repo — .continuity/decisions.json, committed alongside your code

No cloud. No vendor memory. You own the state.

Core Features

Getting Started

1. Install from the VS Code Marketplace
2. Open a project folder in VS Code
3. Choose your profile — the sidebar shows a profile picker on first launch
4. (Optional) Configure MCP — for richer tool integration with Cursor, Copilot, or Gemini
5. Log your first decision — your AI tools will recover this context next session

14-Day Pro Trial

Every install gets full Pro access for 14 days — unlimited decisions, semantic search, knowledge graph, and all features. No credit card required.

After the trial, your decisions are preserved (read-only). Upgrade to Pro to continue logging.

Pricing

Pro Plans:

• Monthly: $9/month
• Annual: $89/year (save 18%)
• Lifetime: $199 one-time

Upgrade to Pro

Works With

All tools work via instruction files. MCP provides additional features for tools that support it.

MCP Tools (59 tools)

Continuity exposes a comprehensive MCP toolset organized into categories:

Decision Management — log_decision, search_decisions, update_decision, list_all_decisions, file_query_result, archive_decisions

Source Ingestion — ingest_source, list_sources, get_source

Reports & Analysis — generate_report (summary, comparison, timeline, deep-dive)

Wiki Health — lint_wiki, get_echo_chamber_report, check_governance, audit_decision_coverage

Context & Session — get_quick_context, read_session_notes, update_session_notes, set_session_subject

Code Intelligence — find_symbol, get_code_links, link_decision_to_code, check_refactoring_safety

Knowledge Graph — get_impact_radius, get_tag_cloud, get_known_subjects

Domain Profiles

Continuity adapts to your domain. Each profile tunes entity detection, search weights, and onboarding:

Select your profile from the sidebar on first launch, or switch anytime via the status bar or Command Palette (Continuity: Select Domain Profile).

What's New

v2.20.18 — April 2026

• Decision Freshness Scoring — New staleness detection engine computes a freshness score (0–100) for every decision based on age, file references, supersession chains, tag activity, and status decay. Stale decisions are automatically annotated in search_decisions and get_quick_context results when they exceed the configurable threshold (default 30).
• Freshness Report Tool — New get_freshness_report MCP tool surfaces the most stale decisions and suggests Dream consolidation candidates.
• VS Code Setting — continuity.freshnessThreshold lets users tune the staleness threshold from the Settings UI.
• Instruction File Integration — Generated instruction files (CLAUDE.md, AGENTS.md, etc.) now include a Decision Freshness section with stale-decision guidance.

v2.20.16 — April 2026

• Search Precision Hardening — Reciprocal Rank Fusion (RRF) stabilizes hybrid search by merging semantic and keyword ranks. Score-Gap Adaptive Limiting ("Cliff Detection") auto-prunes low-confidence results. Redundancy deduplication filters semantically duplicate decisions (cosine similarity > 0.80). Lead-sentence specificity boost rewards decisions that directly address the query topic.
• Peer-Review Benchmark Suite — Project Chronos benchmark suite with 5 benchmarks scored by live Claude Haiku 4.5 API (zero mocked scores). Frameworks: BEIR/TREC retrieval, RAGAS generation quality, RGB robustness, token scaling, ANN vector performance. Results: 0.68 context precision, 0.91 faithfulness, 0.95 robustness, 98.1% token savings at 5K decisions.
• Scholarly Evaluation Document — Full academic paper (benchmarks/ACADEMIC_SUMMARY.md) with 7 references (NeurIPS, EACL, AAAI), comparison against published baselines, threats to validity, and reproducibility instructions.
• Build Hygiene — Benchmarks directory excluded from tsconfig, webpack, and .vscodeignore to prevent build pollution.

v2.20.14 — April 2026

• 7 New CLI Commands — update, archive, lint, report, source, audit, echo-chamber. All critical features previously MCP-only now have CLI equivalents.
• Settings Tab Cleanup — MCP section replaced with small "Advanced" area. Header MCP dot and label removed. Onboarding buttons no longer push MCP setup.
• Instruction Files Rewritten — Generated CLAUDE.md, AGENTS.md, etc. no longer reference MCP tool calls. CLI commands and grep are primary.
• Pre-commit Hook Improved — Reduced false positive warnings for sidebar case handlers.

v2.20.12 — April 2026

• MCP Demoted to Optional — MCP is no longer treated as required. No more auto-setup on activation, disconnect warnings, zombie cleanup intervals, or setup banners. Instruction files (CLAUDE.md, .cursorrules, etc.) are now the primary context delivery mechanism. MCP remains available for users who want it.
• Removed Cline/Roo Integration — Cline and Roo Code MCP configuration removed. Supported tools: Claude Code, Cursor, GitHub Copilot, Google Gemini, OpenAI Codex, Google Antigravity.
• Clean Status Bar — No more MCP warning icon. Status bar shows domain emoji + decision count only. MCP status in tooltip only.
• MCP Dot Indicator Fixed — Sidebar MCP dot was hardcoded green. Now reflects real connection status (hidden when disconnected, green when connected).

v2.20.8 — April 2026

• Persistent MCP (HTTP mode) — One-click command (Continuity: Enable Persistent MCP) starts the MCP server as a daemon on port 3100. Survives VS Code restarts, eliminates zombie processes. Uses StreamableHTTPServerTransport from the MCP SDK.
• MCP Settings Panel — Transport mode (auto/stdio/http), HTTP port, enable/disable persistent MCP, daemon status check, and tool profile selector all in the sidebar Settings tab.
• Autolog Hardening — Self-healing decision storage, startup recovery queue, hook durability improvements, CLI fallback behavior fixes, and repair flow robustness across 5 merged hardening branches.
• Type Unification — Cleaned up MCP server handler types, removed dead code, fixed syntax error that broke esbuild.

v2.20.6 — April 2026

• Security & Governance Dictionaries — Comprehensive security terms added to all 5 entity dictionaries: authentication (SAML, SSO, OIDC, LDAP, RBAC, ABAC), encryption (TLS, AES, RSA, HMAC), vulnerabilities (CSRF, XSS, SQL injection, CVE, OWASP), compliance (GDPR, HIPAA, SOC2, PCI-DSS, ISO 27001), and infrastructure security (firewall, WAF, VPN, zero-trust, Vault, IAM).
• Governance Conflict Detection — Security-specific conflict keywords (downgrade, weaken, bypass) and domain contradiction maps for auth, encryption, and compliance. lint_wiki now flags security decisions lacking documented alternatives.
• Compliance Graph Category — New "Compliance" category in the knowledge graph with dedicated color and tag grouping.

v2.20.5 — April 2026

• Fix: CLI tag extraction — bcrypt, argon2, scrypt, pbkdf2, jwt, and oauth are now extracted as tags by the CLI log command (were only in MCP entity dictionaries, not the CLI keyword list).
• Fix: DecisionLogger TypeError — Fixed entities.tools.length crash with optional chaining.
• Unified Status Bar — Consolidated 4 separate status bar items into 1 compact item. Domain emoji + decision count + MCP warning. Full details in tooltip.

v2.20.3 — April 2026

• Zombie MCP Prevention — Hardened heartbeat lock files, faster stdin-close detection, aggressive cleanup of orphaned MCP server processes. Merged 5 targeted fix branches (heartbeat-fixer, pipe-hardener, shutdown-marshal, orphan-reaper, zombie-tester).
• CLI Improvements — sync --silent fully suppressed, domain recognized as config key, graceful error when dashboard is missing the blessed dependency.
• Entity Extraction — Added bcrypt, argon2, scrypt, and pbkdf2 to entity dictionaries for better security-domain coverage.
• ServiceContainer Refactoring — Cleaner dependency injection and updated test coverage.

v2.19.9 — April 2026

• LLM Wiki Pattern — Continuity is now a full persistent, compounding knowledge base with source ingestion, report generation, lint checks, and query write-back.
• Raw Sources Layer — New ingest_source, list_sources, get_source tools. Drop files in .continuity/sources/ and ingest them with entity extraction and deduplication.
• Report Generation — New generate_report tool with 4 formats: summary, comparison table, timeline, and deep-dive analysis.
• Wiki Lint — New lint_wiki tool: 7 health checks producing a 0-100 health score.
• Query Write-Back — New file_query_result tool files search syntheses as wiki-page decisions with reciprocal relationships.
• Graph Visualization — Source document nodes, orphan highlighting, stale indicators, color-coded edges, legend.
• Domain Profile Picker — Visual sidebar card with mascot avatars. Status bar shows domain with QuickPick switcher.

v2.19.4 — April 2026

• Anti-Sycophancy Guardrails — Detects and breaks echo chambers in the decision pipeline. Based on research showing AI sycophancy + persistent memory compounds false confidence.
• Epistemic Rigor Scoring — New quality dimension evaluating whether decisions document alternatives and trade-offs.
• Contradiction-Aware Search — Search results now surface tensions between contradicting decisions.
• Echo Chamber Report — Analyzes the decision corpus for bias patterns.
• Review Decay — Notifications for high-priority decisions that haven't been reviewed.

v2.19.0 — April 2026

• Domain Profiles — 6 domains: software, writing, research, medical, legal, general. Each with tailored entities, search weights, and onboarding.
• Engagement Suite — Onboarding cards, memory score, contextual nudges, value messages.
• Graph Improvements — Entity details on selection, 12 category system, progressive rendering.
• Priority Inference — Decisions auto-prioritized by content analysis.

v2.17.3 — April 2026

• Lazy MCP Startup — Server responds immediately, defers expensive work to background.
• Entity-Graph Search — search_decisions supports local, global, and hybrid graph-traversal modes.
• Citation Sources — Search results include commit hashes and file references for provenance.

Privacy

• All data stays local — decisions, sources, and session notes live in .continuity/ in your workspace
• No cloud storage — Continuity never sends your decisions to any server
• Anonymous analytics — opt-out via continuity.analytics.enabled: false in VS Code settings
• API keys stored securely — uses VS Code SecretStorage (encrypted, never plain text)

Support

• Issues: GitHub
• Email: support@hackerware.com
• Website: hackerware.com