# Denpex MCP server Agent View of the PolicyLayer registry record for Denpex: identity, probed posture, risk grade, and all 15 tools classified. HTML page: https://policylayer.com/tools/denpex-mcp ## Facts - Server id: `denpex-mcp` - Install: `npx -y denpex-mcp` - Homepage: https://www.npmjs.com/package/denpex-mcp - Registry record: grade F, identity unverified - Lifecycle: active - Rate-limited: no - Tools: 15 (11 with introspected schema) - Tool categories present: Execute, Read, Write - Context-window cost: 2718 tokens per request — https://policylayer.com/token-cost/denpex-mcp - Tags: denpex mcp, automation - Record last modified: 2026-08-05T21:20:27.351Z ## Tools (15) | Tool | Category | Risk | Record | | --- | --- | --- | --- | | `compare_runs` | Execute | High | https://policylayer.com/tools/denpex-mcp/compare-runs.md | | `diff_environments` | Execute | High | https://policylayer.com/tools/denpex-mcp/diff-environments.md | | `node_health` | Execute | High | https://policylayer.com/tools/denpex-mcp/node-health.md | | `predict_failures` | Execute | High | https://policylayer.com/tools/denpex-mcp/predict-failures.md | | `analyze_cascade` | Read | Low | https://policylayer.com/tools/denpex-mcp/analyze-cascade.md | | `ask_followup` | Read | Low | https://policylayer.com/tools/denpex-mcp/ask-followup.md | | `check_quota` | Read | Low | https://policylayer.com/tools/denpex-mcp/check-quota.md | | `check_versions` | Read | Low | https://policylayer.com/tools/denpex-mcp/check-versions.md | | `diagnose_crash` | Read | Low | https://policylayer.com/tools/denpex-mcp/diagnose-crash.md | | `list_incidents` | Read | Low | https://policylayer.com/tools/denpex-mcp/list-incidents.md | | `lookup_failure` | Read | Low | https://policylayer.com/tools/denpex-mcp/lookup-failure.md | | `preflight_check` | Read | Low | https://policylayer.com/tools/denpex-mcp/preflight-check.md | | `recent_diagnoses` | Read | Low | https://policylayer.com/tools/denpex-mcp/recent-diagnoses.md | | `validate_checkpoint` | Read | Low | https://policylayer.com/tools/denpex-mcp/validate-checkpoint.md | | `submit_fix_outcome` | Write | Medium | https://policylayer.com/tools/denpex-mcp/submit-fix-outcome.md | ## Tool descriptions - `compare_runs` — 'It worked yesterday.' Give it the logs from the run that worked and the run that failed, and it isolates what actually changed between them instead of re-diagnosing the failure from scratch. Use this whenever a previously-green job star… - `diff_environments` — Diff the SOFTWARE environment between a good run and a bad one — torch / CUDA / NCCL / driver / flash-attn versions, env vars — and flag combinations known to be broken. Feed it pip freeze, nvidia-smi, env output (or the whole job log; i… - `node_health` — Score the health of every node Denpex is monitoring in your fleet, worst first. Use it to answer 'which nodes should I exclude from this job?' before scheduling, or 'what else is degrading?' after an incident. Scale+ feature; requires th… - `predict_failures` — List the predictive risk scores Denpex has computed for your fleet — nodes whose counters are trending toward a failure that has not happened yet (rising correctable ECC, row-remap pressure, PCIe replays, thermal drift). Use it to drain … - `analyze_cascade` — Analyze a multi-rank distributed-training failure: find the rank that failed FIRST (the others usually just time out waiting on it) and route any GPU Xid faults to the correct node action (reset / reboot / RMA). Runs offline and instantl… - `ask_followup` — Ask the Denpex diagnostician a follow-up question about a crash — clarify the fix, explore an alternative, or dig deeper. Pass the original logs and your question; pass back the conversationId from a previous reply to continue the thread… - `check_quota` — Show your Denpex plan and how many diagnoses you have left this month. - `check_versions` — Scan logs (or an environment dump like pip list / nvidia-smi) for CUDA / PyTorch / cuDNN / NCCL / driver versions that are known to be incompatible or buggy together, and get the recommended pin. Requires the Team plan or an active trial. - `diagnose_crash` — Diagnose a GPU/ML training crash (CUDA OOM, NCCL timeout, Xid fault, NaN loss, checkpoint corruption, etc.) and get the root cause plus an escalating fix (primary → fallback → emergency). Provide ONE of: logs (paste), file (path to a log… - `list_incidents` — List the GPU/training incidents Denpex's always-on monitor caught and auto-escalated on your fleet — failure type, cluster, escalation level, and the fix it applied. Requires the Scale plan or higher. - `lookup_failure` — Search Denpex's bundled failure encyclopedia by error code, keyword, or symptom (e.g. 'Xid 79', 'NCCL timeout', 'loss is NaN', 'CUDA out of memory'). Returns the most relevant known failures with their fixes. Works fully offline and cost… - `preflight_check` — Answer 'is this cluster safe to launch on?' BEFORE you burn GPU-hours on a job that will die in the first ten minutes. Feed it the health artifacts from the nodes you are about to use — nvidia-smi -q, ibstat, nvidia-smi topo -m, dmesg -T… - `recent_diagnoses` — Show your most recent Denpex diagnoses — job name, failure type, confidence, and the fix — so you can recall what a prior crash was and how it was resolved. - `validate_checkpoint` — Before you resume a crashed job, check that the checkpoint you are about to resume FROM is not the one the crash corrupted. Give it ls -la of the checkpoint directory, and optionally sha256sum -c output and any load errors. Returns which… - `submit_fix_outcome` — Report back whether a Denpex fix actually resolved the failure. Call this after you apply a fix from diagnose_crash and observe the result — it is the ONLY way the engine learns, and it directly improves the next diagnosis of this failur… ## Related servers - UnClick (1662 tools) — https://policylayer.com/tools/io-github-malamutemayhem-unclick-mcp-server.md - Delx Mcp A2a (1076 tools) — https://policylayer.com/tools/io-github-davidmosiah-delx-mcp-a2a.md - Delx Protocol — Agent Recovery & Continuity (1076 tools) — https://policylayer.com/tools/io-github-davidmosiah-delx-protocol.md - Nodebench (824 tools) — https://policylayer.com/tools/io-github-homenshum-nodebench.md - MCP Framework Personal (790 tools) — https://policylayer.com/tools/inggerman-mcps.md - Delx MCP Server (740 tools) — https://policylayer.com/tools/io-github-davidmosiah-delx-mcp-server.md - Binance MCP Server (734 tools) — https://policylayer.com/tools/nirholas-binance-mcp.md - Crawlora (728 tools) — https://policylayer.com/tools/crawlora-mcp.md ## For agents This record is a snapshot. Live verdicts and the full registry: - Check every server in your MCP config at once: `npx -y policylayer stack` - Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md) - Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events` --- Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=denpex-mcp · API: https://policylayer.com/registry/api · Recommended policies for every tool: https://policylayer.com/policies/denpex-mcp