registry  /  @splyntra/sdk  /  1.2.0

@splyntra/sdk@1.2.0

Splyntra TypeScript SDK - Agent observability & security, built on OpenTelemetry

AI Security Review

scanned 10d ago · by lpm-firewall-ai

No confirmed malicious attack surface. Runtime telemetry export is package-aligned observability behavior and requires user configuration or invocation.

Static reason
One or more suspicious static signals were detected.
Trigger
User imports package and constructs Splyntra or calls instrumentation helpers.
Impact
Telemetry metadata may be sent to the configured collector endpoint as documented; no unauthorized persistence, code execution, credential harvesting, or agent control-surface mutation found.
Mechanism
OpenTelemetry span creation/export and optional SDK method wrapping
Rationale
Static inspection shows a normal OpenTelemetry-based observability SDK with explicit initialization, configurable collector endpoint, and built-in redaction. Scanner secret findings are synthetic test fixtures for redaction behavior, not evidence of credential exposure or malware.
Evidence
package.jsondist/index.jsdist/exporters/index.jsdist/redaction.jsdist/instrumentors/index.jsdist/instrumentors/openai.jsdist/instrumentors/langgraph.jsdist/instrumentors/crewai.jsdist/instrumentors/openai-agents.jsdist/redaction.test.js
Network endpoints1
localhost:4318/v1/traces

Decision evidence

public snapshot
AI called this Clean at 94.0% confidence as Benign with low false-positive risk.
Evidence for block
    Evidence against
    • package.json has no install/postinstall hooks; only prepublishOnly build for publisher workflow.
    • dist/index.js requires explicit Splyntra construction with apiKey/project before exporter setup.
    • dist/exporters/index.js sends OpenTelemetry traces only to caller-provided endpoint, defaulting to localhost /v1/traces.
    • dist/redaction.js redacts AWS/JWT/bearer/API-key patterns before export when default processor is enabled.
    • dist/instrumentors/*.js only monkey-patch known SDK methods when user invokes instrumentation; no fs, child_process, eval, or agent config writes found.
    • dist/redaction.test.js contains synthetic example secrets used to test redaction, not live credentials.
    Behavioral surface
    SourceNo risky source behavior triggered.
    Supply chain
    HighEntropyStrings
    ManifestNo manifest risk signals triggered.
    scanned 9 file(s), 31.7 KB of source

    Source & flagged code

    3 flagged · loading source
    dist/redaction.test.jsView file
    8patternName = aws_access_key severity = critical line = 8 matchedText = const ou...d");
    Critical
    Critical Secret

    Package contains a critical-looking secret pattern.

    dist/redaction.test.jsView on unpkg · L8
    8patternName = aws_access_key severity = critical line = 8 matchedText = const ou...d");
    Critical
    Secret Pattern

    AWS access key ID in dist/redaction.test.js

    dist/redaction.test.jsView on unpkg · L8
    27patternName = aws_access_key severity = critical line = 27 matchedText = const sp...} };
    Critical
    Secret Pattern

    AWS access key ID in dist/redaction.test.js

    dist/redaction.test.jsView on unpkg · L27

    Findings

    3 Critical3 Low
    CriticalCritical Secretdist/redaction.test.js
    CriticalSecret Patterndist/redaction.test.js
    CriticalSecret Patterndist/redaction.test.js
    LowNon Install Lifecycle Scripts
    LowScripts Present
    LowHigh Entropy Strings