Half the tokens.
All the context.
gotcontext is the context gateway for AI agents. Semantic compression, code intelligence, security scanning and team memory, behind one MCP URL.
Defining your context gateway
Claude, GPT, Gemini, local models. Compression happens before your tokens leave, so every provider benefits equally.
One Streamable HTTP URL with a bearer token. Your agent discovers every tool the moment it connects.
Semantic compression runs on our metal with ONNX embeddings. Your context is never forwarded to an AI provider.
Your context goes in. Half the tokens come out.
Model vendors bill by the token. We shrink the bill.
The engine builds a semantic skeleton of what you were about to send, keeps the meaning and drops the rest. Ratios below are measured and locked by tests, not marketing.
Point your agent at code. Get answers, not file dumps.
Blast radius, callers, edit plans and 128 AST security rules run server side, so your agent reads one ranked answer instead of twenty raw files.
$ gc_blast_radius record_usage ├─ blast_radius_score moderate ├─ caller_tree │ ├─ routers/usage.py │ ├─ mcp_gateway.py │ └─ services/billing.py ├─ impacted tests 6 files └─ graph_trust_summary parser-backed
A knowledge base your whole agent fleet shares.
Agents write once and every agent reads compressed. Plans are versioned items with signed change proposals only a human owner can merge.
Watch it compress. This one is real.
Built for the enterprise from day one
Run the full gateway in your VPC from one Docker image, licensed with Ed25519-signed keys.
Clerk-backed sessions and JWTs for the dashboard, HMAC-signed gc_ keys for machines.
Per-project budgets with 75, 90 and 100 percent alerts, usage rollups by key and model.
Append-only audit events enforced at the database layer. Nothing is silently rewritten.
A Linux Foundation Agent2Agent card, signed share manifests and cross-agent task delegation.
gc_scan runs 128 AST rules across 6 packs. gc_skill_scan gates skills before your agents install them.
For every agent stack that pays for context
Ready to cut your token bill?
Start free with 23 MCP tools. Connect one URL. Keep your models.
Compatible with
Compress it. Remember it. Delegate it. Scan it.
One bearer token runs all five tool families below. Same key, same Authorization header.
Compress
Every tool response shrinks before your agent reads it, via ingest_context, read_skeleton, and search_semantic.
Try it below →Remember
A knowledge base any of your agents can query, via gc_kb_query and the rest of the gc_kb_* family. Needs a project-bound key.
Read the docs →Delegate
Hand a task to a peer agent and review its plan before it merges, via gc_a2a_task_create and gc_plan_propose.
See the agent card →Scan
gc_scan runs 128 AST rules across 6 packs on your code. gc_skill_scan returns a safe_to_install verdict on a SKILL.md or MCP manifest.
See how it works →Measure
An open, runnable leaderboard for inference cost and speed. No account needed to read it.
View the leaderboard →Why the output is auditable.
Not a summary. A re-ranking of your own sentences. Documents are chunked, embedded, and scored on a semantic graph; only the highest-ranked nodes survive into the output. Typically ~50% smaller on production traffic. Methodology & benchmark peak in the measurement section. measurement section.
- Three compression modes: fast / balanced / SBERT
- AST-aware code compression for 7+ languages
- Per-workspace key scoping. Keys cannot read across workspaces.
- Command Palette: Cmd+K navigation, G+D shortcuts, full-text search
- GitHub Integration: token-savings summaries posted on your pull requests
- Real-Time Queue Monitor: live SSE streaming for batch jobs
- Roles: Owner, Admin, Member, Viewer. Shared projects, activity feed.
- Stacks with native prompt caching (Anthropic / OpenAI / Gemini). When both apply, total input-cost reduction can reach 95%. See methodology.
How a response gets compressed.
Same input, same output, every run. Four steps, no model in the loop. The output is a re-ranking of your own sentences. Every token in the compressed response appears in the original.
140+ MCP tools behind one endpoint.
Works with any MCP-compatible client. Claude Code, Codex, Gemini CLI, Cursor, VS Code. One command configures the MCP server. No JSON editing required.
- CWE-22 path traversal prevention on all file I/O
- Async batch ingest: 4× throughput
- Prometheus metrics, OpenTelemetry tracing, health checks
gc_compress_manifestshrinks MCP tool-description bloat
- 1. Get a free key from the dashboard
- 2. Run the CLI. It prompts for your key
- 3. Restart your CLI
$ npx gotcontext wrap claudenpx gotcontext doctor — shows which CLIs are detected and configured.Prefer manual JSON config?
{
"mcpServers": {
"gotcontext": {
"url": "https://api.gotcontext.ai/mcp",
"headers": {
"Authorization": "Bearer gc_your_key_here"
}
}
}
}The tools chain into one loop.
This is the same loop we run against our own codebase: one document gets ingested, compressed, stored in Knowledge, handed to another agent, and merged back in, all under one key.
ingest_contextread_skeletongc_kb_querygc_a2a_task_creategc_plan_decideTry it now
Paste any text and see how much you can save. No signup required.
Text is processed in-memory and is not stored, logged with PII, or used for training. Do not paste secrets or production credentials. Privacy details →
Your codebase is the biggest thing your agent reads.
Text summarizers compress prose. We compress code, AST-aware and structure-preserving, at 10-11× on real source files. An agent that reads mcp_gateway.py in full spends 20,076 words of context. With read_skeleton it gets a faithful structural skeleton for 1,935 words and drills into any function on demand. Same answerable questions, one-tenth the context budget.
Compression is size-dependent. Small files compress little. The engine keeps them faithful. The ratios above are on large files where agents actually struggle to fit the full source into context.
Two tools, one MCP session.
Point either one at your own content. Both run through the same Authorization: Bearer gc_ header as everything else on this page.
gc_scan128 AST rules across 6 packs run against your code: auth, crypto, deserialization, secrets, subprocess, tls-safe.
gc_skill_scanReads a SKILL.md or MCP tool manifest and returns a safe_to_install verdict before you install it.
Pay for tool calls. Compression is included.
Every MCP tool response is compressed before it returns to your agent, so each call delivers more context per token. The multiplier scales with the live compression ratio (see hero). Covers solo developers to enterprise teams.
Free
- 1,000 compressions/month
- 100KB max document
- Standard compression
- Command Palette & shortcuts
- Activity Feed
- Dark/Light theme
- Community support
Pro
- 50,000 compressions/month
- All 140+ MCP tools (incl. ACE, knowledge mgmt, multimodal)
- Priority queue: 2 concurrent compression slots
- 1MB max document
- Accelerated compression (3-5x faster)
- Queue Monitor (real-time SSE)
- Usage analytics
- Webhook Notifications
- Priority support
Business
- 500,000+ compressions / month
- All 140+ MCP tools
- Priority queue: 8 concurrent compression slots
- Self-hosted Docker (run in your VPC)
- OIDC federation (Okta, Auth0, Azure AD)
- Audit-log export (NDJSON/CSV) for compliance reviews
- SBERT embeddings (higher fidelity than the default MiniLM tier)
- SSO / SAML
- Email support · SLA on request (custom MSA)
- DPA / IP indemnity / custom MSA
Start free.
1,000 compressions/month, all 140+ tools, no credit card.