BaseCode – MDM for Coding Agents

sivasurend1 pts1 comments

basecode — MDM for Coding Agents

Talk to us<br>Get started<br>Try demo

Observe<br>Govern<br>Adopt<br>Platform<br>Pricing<br>Log in

Talk to us<br>Try demo

OBSERVE EVERYTHING · ENFORCE ANYTHING

Know and control how coding agents are being used in your organization.

Your engineers run Claude Code, Codex, Copilot, Gemini and opencode on their machines. basecode lets you centrally inject agent identities, coding policies, skills, MCP servers and models into all of them — then observe every prompt, project and token , and enforce guardrails and budgets across the fleet.

Talk to us<br>Try demo →

claude-code<br>codex<br>copilot<br>gemini<br>opencode<br>architect BUILT-IN

company policy, enforced by the agent

profile: architect-adk · synced from gateway

▲ basecode<br>architect-adk · s1 · 33:54

Ctrl-] detach (keeps running) · F2 sidebar

⚠ 3 MCP servers need authentication · run /mcp

can you build a langraph agent

● No — this agent only builds with the Lyzr Agent Development Kit (lyzr‑adk), not LangGraph. If you tell me what you're actually trying to build, I can map it to Lyzr ADK equivalents:

LangGraph conceptLyzr ADK equivalent<br>Graph nodes/edges for multi-step flowMultiple studio.create_agent() instances, orchestrated via sequential .run() calls<br>Shared state across nodesContext objects or session memory<br>Tool-calling nodesagent.add_tool() with typed Python functions<br>Persistence/checkpointingmemory=N + session_id on agent.run()

A developer asks Claude Code for a LangGraph agent. The architect-adk profile — pushed by their admin — makes the agent refuse and redirect to the company-approved stack. No plugin. No fork. Just policy.

100%

Of traffic observed & audit-logged

6+

Coding agents managed

Any

Model provider — hosted or self-hosted

Plugins or forks required — 100% additive

Employees leave. Knowledge stays.

Every project's context, decisions and learnings are captured in a knowledge graph on your gateway — not in one person's head. Attrition doesn't take your codebase's memory with it.

Nothing is a black box.

Every prompt, token, model call and tool call across the fleet is observed, attributed and audit-logged — so what your agents build is never a mystery.

Day-one onboarding.

A new hire's agent inherits profiles, skills, MCP servers and project knowledge the moment their device enrolls — they build like a veteran from the first prompt.

additive by design·works with any IDE ·<br>syncs live over wss:// ·enforced at the gateway ·<br>OPA / Cedar guardrails·no plugin · no fork · just policy ·

shreyas-mac · projects<br>53 projects · 123h 47m · 230.2M tok

harness-agent-service 53h 21m · 97.3M tokcccdxcplt

flabot 23h 43m · 63.0M tokcc

gitview 6h 35m · 23.1Mcc

vite-architect-core 5h 06m · 10.2M tok

hyoconpute 2h 47m

glupgorton 14h 03m · 36.3M tokcdx

architect-studio 4h 47m

desktop 1h 13m

agent-studio-ui 2h 11m

proxy 58m

docs 41m

3 Owners

34 Devices

7d 29h Coding time

601.2M Tokens

tile size = total coding time · duplicate device snapshots counted once

SEE — 01 · FLEET OBSERVABILITY<br>Who is building what — across every person, project and device.

basecode attributes real coding activity — active coding time, recency and token spend — across Claude Code, Codex and Copilot , rolled up per person or per group. Duplicate device snapshots of the same folder are counted once, across all of a person's devices.

✓Per-person or per-group treemaps — pivot by coding time, 7-day activity, or token spend.

✓Agent attribution — every project shows which agents built it: cc, cdx, cplt badges.

✓Zoom out to the whole org — regions group the same project across owners; click a tile to open that person.

llm logs · last hour<br>LIVE

1,284 Requests

2,845ms Avg latency

134.8k Tokens

US$0.14 Cost

13:12:09RESPONSES STREAM[SUGGESTION MODE: Suggest what the user mig…3.2s

13:11:55RESPONSES STREAMclear9.2s

13:11:22DENIED · OPAprompt contained customer PII — redacted in-line0ms

13:10:31RESPONSES STREAMAvailable agent types for the Agent tool: …8ms

12:51:11MCP TOOL CALLgithub.create_pull_request — audit-logged0.6s

SEE — 02 · GUARDRAILS & LLM LOGS<br>See every prompt. Guard every response.

Every request from every employee's coding agent flows through the gateway — and shows up in a live log with latency, tokens and cost. Enforcement runs in the request path, at the gateway — it can't be switched off from a laptop — and every decision is retained as an audit-ready record.

✓Policy-engine guardrails — evaluate prompts and responses against org policy with OPA / Cedar: allow, deny or redact before anything reaches a model.

✓Slice by anything — status, model, provider, virtual key, routing rule, user, team or customer.

✓MCP logs too — every connector interaction observed and audit-logged.

model routing<br>per-agent · fully additive

claude-codecodexcopilotopencode

↓ ↓ ↓ ↓

▲ basecode gateway

↓ ↓ ↓ ↓

OpenAIAnthropicVertex AIvLLMFireworksyour GPUs

403 Model not allowed — scoped virtual key

COST — 03–04 · BUDGETS & MODEL GATEWAY<br>Remotely swap the...

agent coding across gateway model basecode

Related Articles