Sdl — the complete, unedited output of the deterministic mcptrustchecker engine v1.13.0, scanned . Every finding, capability tag and score component below is exactly what the engine produced — no AI, no post-processing.
{
"tool": {
"name": "mcptrustchecker",
"version": "1.13.0",
"methodologyVersion": "mcptrustchecker-1.13"
},
"target": {
"id": "sdl-mcp",
"source": {
"kind": "package",
"origin": "sdl-mcp"
},
"server": {
"name": "sdl-mcp"
}
},
"grade": "A",
"score": {
"score": 92,
"threatScore": 99,
"grade": "A",
"band": "A",
"categorySubtotals": {
"injection": 0,
"exfiltration": 0,
"permissions": 0,
"supply-chain": 1.2,
"network": 0,
"hygiene": 0
},
"vector": [
{
"kind": "threat",
"ruleId": "MTC-SUP-010",
"category": "supply-chain",
"severity": "low",
"confidence": "heuristic",
"rawWeight": 3,
"confidenceMult": 0.4,
"diminishingFactor": 1,
"appliedPenalty": 1.2
},
{
"kind": "client",
"term": "capability-exposure",
"level": "high",
"label": "capability blast radius (high) — client exposure if the model is manipulated",
"appliedPenalty": 6
},
{
"kind": "client",
"term": "verification-discount",
"level": "repo",
"label": "publisher verification (public source) — no provenance, but the source is public and inspectable",
"appliedPenalty": 1
},
{
"kind": "client",
"term": "coverage-honesty",
"level": "source",
"label": "inspection depth (source) — how much of the target the scan could see",
"appliedPenalty": 0
}
],
"gatesFired": [],
"methodologyVersion": "mcptrustchecker-1.13"
},
"capability": {
"level": "high",
"reasons": [
"reads sensitive or local data",
"has a read → egress path (a data-exfiltration surface)",
"can execute shell commands or code"
],
"tags": [
"sensitive-source",
"code-exec"
]
},
"coverage": {
"level": "source",
"inputs": {
"toolSurface": true,
"implementationSource": true,
"packageMetadata": true,
"liveTransport": false
},
"caveats": [
"Tools were statically extracted from the published source (40 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command \"npx -y <package>\"."
]
},
"findings": [
{
"ruleId": "MTC-SRC-006",
"title": "Reads a sensitive credential path or dumps the environment (dist/mcp/tools/search-edit/planner.js)",
"category": "exfiltration",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`dist/mcp/tools/search-edit/planner.js:210`): Reading private keys / cloud credentials, or serializing the whole environment, is a sensitive-data source that becomes exfiltration when combined with any egress. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/mcp/tools/search-edit/planner.js"
},
"evidence": "ENYLIST = new Set([ \".npmrc\", \".netrc\", \".pgpass\", \".my.cnf\", \".boto\", ]); const SECRET_FILENAME_RE",
"owasp": "LLM02:2025 Sensitive Information Disclosure",
"data": {
"rule": "MTC-SRC-006",
"file": "dist/mcp/tools/search-edit/planner.js",
"line": 210,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (dist/cli/setup-wizard/lsp.js)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`dist/cli/setup-wizard/lsp.js:1`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/cli/setup-wizard/lsp.js"
},
"evidence": "spawnSync } from \"node:child_process\"; import { platform } from \"node:os\"; const LSP_REGISTRY = { ts: { lan",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "dist/cli/setup-wizard/lsp.js",
"line": 1,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (dist/db/ladybug-algorithms.js)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`dist/db/ladybug-algorithms.js:54`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/db/ladybug-algorithms.js"
},
"evidence": "try { await exec(conn, \"INSTALL algo\"); } catch (installErr) { logger.debug(\"lad",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "dist/db/ladybug-algorithms.js",
"line": 54,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (dist/db/ladybug-clusters.js)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`dist/db/ladybug-clusters.js:19`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/db/ladybug-clusters.js"
},
"evidence": "(conn, row) { await exec(conn, `MATCH (r:Repo {repoId: $repoId}) MERGE (c:Cluster {clusterId: $clusterId})",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "dist/db/ladybug-clusters.js",
"line": 19,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (dist/db/ladybug-config.js)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`dist/db/ladybug-config.js:3`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/db/ladybug-config.js"
},
"evidence": "(conn, row) { await exec(conn, `MERGE (p:ToolPolicyHash {policyHash: $policyHash}) SET p.policyBlob = $policyBl",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "dist/db/ladybug-config.js",
"line": 3,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (dist/db/ladybug-core.js)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`dist/db/ladybug-core.js:657`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/db/ladybug-core.js"
},
"evidence": "})); } export function exec(conn, statement, params = {}) { const checkpointError = rawCheckpointError(statement);",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "dist/db/ladybug-core.js",
"line": 657,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (dist/db/ladybug-derived-state.js)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`dist/db/ladybug-derived-state.js:57`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/db/ladybug-derived-state.js"
},
"evidence": "onn) => { await exec(wConn, \"MERGE (d:DerivedState {repoId: $repoId}) ON CREATE SET d.clustersDirty = $clustersD",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "dist/db/ladybug-derived-state.js",
"line": 57,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (dist/db/ladybug-edges.js)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`dist/db/ladybug-edges.js:80`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/db/ladybug-edges.js"
},
"evidence": "N_FILE joins. await exec(conn, `MATCH (r:Repo {repoId: $repoId}) MERGE (a:Symbol {symbolId: $fromSymbolId})",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "dist/db/ladybug-edges.js",
"line": 80,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (dist/db/ladybug-embeddings.js)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`dist/db/ladybug-embeddings.js:11`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/db/ladybug-embeddings.js"
},
"evidence": "(conn, row) { await exec(conn, `MERGE (e:SymbolEmbedding {symbolId: $symbolId}) SET e.model = $model,",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "dist/db/ladybug-embeddings.js",
"line": 11,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (dist/db/ladybug-feedback.js)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`dist/db/ladybug-feedback.js:4`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/db/ladybug-feedback.js"
},
"evidence": "(conn, row) { await exec(conn, `MERGE (a:Audit {eventId: $eventId}) SET a.timestamp = $timestamp, a.to",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "dist/db/ladybug-feedback.js",
"line": 4,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-005",
"title": "Dynamic module load from a non-literal (dist/indexer/adapter/plugin/loader.js)",
"category": "permissions",
"severity": "medium",
"confidence": "heuristic",
"description": "In the server's implementation (`dist/indexer/adapter/plugin/loader.js:71`): Loading a module chosen at runtime (from a variable) can pull in and run attacker-influenced code paths. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/indexer/adapter/plugin/loader.js"
},
"evidence": "const module = await import(importUrl); const pluginModule = module.default ?? module; if (!pluginMod",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-005",
"file": "dist/indexer/adapter/plugin/loader.js",
"line": 71,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-005",
"title": "Dynamic module load from a non-literal (dist/indexer/language-packs.js)",
"category": "permissions",
"severity": "medium",
"confidence": "heuristic",
"description": "In the server's implementation (`dist/indexer/language-packs.js:618`): Loading a module chosen at runtime (from a variable) can pull in and run attacker-influenced code paths. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/indexer/language-packs.js"
},
"evidence": "return (await import(adapterModule)); } catch (err) { lastErr = err; i",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-005",
"file": "dist/indexer/language-packs.js",
"line": 618,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-005",
"title": "Dynamic module load from a non-literal (dist/indexer/treesitter/grammarLoader.js)",
"category": "permissions",
"severity": "medium",
"confidence": "heuristic",
"description": "In the server's implementation (`dist/indexer/treesitter/grammarLoader.js:127`): Loading a module chosen at runtime (from a variable) can pull in and run attacker-influenced code paths. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/indexer/treesitter/grammarLoader.js"
},
"evidence": "try { return require(packageName); } catch (error) { const packageRoot = grammarPackageRoots.",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-005",
"file": "dist/indexer/treesitter/grammarLoader.js",
"line": 127,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-005",
"title": "Dynamic module load from a non-literal (dist/native/addon-loader.js)",
"category": "permissions",
"severity": "medium",
"confidence": "heuristic",
"description": "In the server's implementation (`dist/native/addon-loader.js:7`): Loading a module chosen at runtime (from a variable) can pull in and run attacker-influenced code paths. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/native/addon-loader.js"
},
"evidence": "didate = (candidate) => require(candidate); const defaultLogFailure = (failure) => { logger.debug(\"Native addon unav",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-005",
"file": "dist/native/addon-loader.js",
"line": 7,
"nonRuntime": false
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in packaging/dev tooling (dist/benchmark/external-runner.js)",
"category": "permissions",
"severity": "low",
"confidence": "heuristic",
"description": "In a packaging/dev/install script (shipped, but not the server runtime) (`dist/benchmark/external-runner.js:1`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
"remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
"location": {
"kind": "server",
"name": "dist/benchmark/external-runner.js"
},
"evidence": "ync, spawn } from \"node:child_process\"; import { createHash } from \"node:crypto\"; import fs, { existsSync } from \"node:f",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "dist/benchmark/external-runner.js",
"line": 1,
"nonRuntime": true
}
},
{
"ruleId": "MTC-SUP-010",
"title": "Package runs install-time scripts (postinstall)",
"category": "supply-chain",
"severity": "low",
"confidence": "heuristic",
"description": "\"sdl-mcp\" executes postinstall script(s) at install time. An install hook runs at install time; most are routine build/setup, but review what it does before trusting it.",
"remediation": "Review the scripts; install with --ignore-scripts where possible and vet what they do.",
"location": {
"kind": "package",
"name": "sdl-mcp"
},
"evidence": "node scripts/postinstall.mjs",
"owasp": "LLM03:2025 Supply Chain",
"references": [
"https://github.com/ossf/package-analysis",
"https://owasp.org/www-project-top-10-ci-cd-security-risks/"
],
"data": {
"scripts": [
"postinstall"
]
}
}
],
"toxicFlows": [],
"capabilities": [
{
"tool": "sdl.action.search",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.manual",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.retrieve",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.workflow",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.context",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.file",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.query",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.code",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.repo",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.agent",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.repo.register",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.repo.status",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.repo.unregister",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.index.refresh",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.buffer.push",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.buffer.checkpoint",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.buffer.status",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.symbol.search",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.symbol.getCard",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.slice.build",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.slice.refresh",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.slice.spillover.get",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.delta.get",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.code.needWindow",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.code.getSkeleton",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.code.getHotPath",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.policy.get",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.policy.set",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.repo.overview",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.usage.stats",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.pr.risk.analyze",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.agent.feedback",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.agent.feedback.query",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.runtime.execute",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.response.get",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.memory.store",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.memory.query",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.memory.remove",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.memory.surface",
"tags": [],
"reasons": {}
},
{
"tool": "sdl.info",
"tags": [],
"reasons": {}
}
],
"surfaceDigest": "c39c9bef52fd4f4d2f7acdd88f624450b3ee1c01f37715488ea2eb7ac2dedf13",
"stats": {
"tools": 40,
"prompts": 0,
"resources": 0,
"findingsBySeverity": {
"critical": 0,
"high": 10,
"medium": 4,
"low": 2,
"info": 0
}
}
}