CI Triage x402
activeDeveloper Reliability Tools
Deterministic CI/CD failure diagnosis, repair planning, test-report parsing, workflow validation, and secret-safe developer utilities paid per call through x402.
- Transactions · 30d
- 0
- Volume · 30d
- $0.00
- Unique buyers · 30d
- 0
- Uptime · 30d
- 100.0%
- Latency p50
- 40ms
Endpoints (30 live)
POST/v1/fingerprint— Generate a stable, secret-free fingerprint from a build or test failure so agents can group duplicate CI errors across changing line numbers, timestamps, paths, identifiers, and retry attempts. (0.001 USDC on Base)POST/v1/sarif-summary— Normalize a SARIF 2.x report from CodeQL, Semgrep, Trivy, or another scanner into severity counts, top rules, prioritized findings, file locations, and concise messages suitable for an agent's remediation queue. (0.006 USDC on Base)GET/v1/github-run— Analyze a public GitHub Actions run URL using failed-job metadata and check annotations. Return the failing job and step, root-cause classification, redacted evidence, and repair guidance without requesting a GitHub token or pretending public raw logs are accessible. (0.005 USDC on Base)POST/v1/coverage-summary— Parse LCOV, Cobertura XML, coverage.py JSON, Istanbul JSON summaries, and common terminal coverage tables. Return line, branch, function, and statement percentages plus threshold status in one normalized response. (0.003 USDC on Base)POST/v1/stack-trace— Parse Python, JavaScript, TypeScript, Java, Go, and Rust stack traces into language, exception, message, and normalized frame objects with function, file, line, and column fields for downstream debugging agents. (0.003 USDC on Base)POST/v1/pytest-summary— Extract the failed pytest node id, assertion or exception, relevant Python frames, likely cause category, focused single-test rerun command, and practical repair steps from a pytest failure log. (0.003 USDC on Base)POST/v1/redact— Strip ANSI and runner noise while redacting GitHub tokens, AWS access keys, bearer credentials, JWTs, API keys, passwords, and common secret assignments from CI/CD logs before an agent stores, shares, or analyzes them. (0.001 USDC on Base)POST/v1/typescript-error— Extract the first actionable TypeScript diagnostic, error code, file, line, and message; group repeated diagnostics and return a focused tsc command plus repair guidance for type, module, configuration, and declaration failures. (0.003 USDC on Base)POST/v1/eslint-summary— Normalize stylish, compact, and common ESLint output into error and warning counts, affected files, first actionable rule violations, focused rerun command, and whether safe automatic fixing may be available. (0.003 USDC on Base)POST/v1/github-actions-lint— Parse a GitHub Actions workflow as inert YAML and report structural errors, missing jobs or runners, malformed steps, dangerous pull_request_target patterns, mutable action references, overly broad permissions, and common expression mistakes. (0.006 USDC on Base)POST/v1/test-report-normalize— Auto-detect JUnit XML, SARIF JSON, LCOV, coverage tables, pytest, Jest, ESLint, Go, Rust, and generic CI output, then return a consistent report type, status, totals, failures, diagnosis, and source-format metadata. (0.005 USDC on Base)POST/v1/rerun-command— Infer a safe, non-executed rerun command from pytest, Jest, Vitest, ESLint, TypeScript, npm, pip, Docker, Go, Rust, Maven, or Gradle failure output so an agent can reproduce the smallest failing unit. (0.002 USDC on Base)POST/v1/jest-summary— Extract the failing Jest suite and test, expected-versus-received assertion, JavaScript or TypeScript frames, focused rerun command, and repair guidance from Jest, Vitest, or compatible test output. (0.003 USDC on Base)POST/v1/failure-cluster— Analyze up to twenty-five build or test logs, group retries and duplicate failures by stable secret-free fingerprints, and return cluster counts, representative messages, failure types, and the highest-frequency root cause. (0.008 USDC on Base)POST/v1/triage— Parse a CI, build, compiler, dependency, container, lint, or test log into a secret-redacted failure type, primary error, useful trace, fix category, and confidence score. Use it to analyze build logs and identify the first actionable root-cause signal. (0.002 USDC on Base)POST/v1/check-annotations— Normalize GitHub Checks or compatible annotation objects into severity counts, affected files, top rules, prioritized evidence, and a concise failure summary. Secret-like content is redacted before it is returned. (0.003 USDC on Base)POST/v1/junit-summary— Parse bounded JUnit XML from pytest, Jest, Maven, Gradle, or other test runners. Return suite totals, pass rate, duration, failed test names, failure messages, and the files or classes that need attention without executing embedded content. (0.004 USDC on Base)POST/v1/npm-error— Identify npm and Node.js failure modes such as ERESOLVE peer conflicts, ETARGET versions, engine mismatches, node-gyp builds, permission errors, missing files, lockfile drift, and registry/network failures, then return safe recovery commands. (0.004 USDC on Base)POST/v1/markdown-handoff— Convert a raw failing build or test log into concise secret-redacted Markdown containing the root cause, evidence, fingerprint, reproduction command, repair checklist, and confidence for an issue, pull request, or human reviewer. (0.004 USDC on Base)POST/v1/cargo-error— Extract Rust compiler error codes, source locations, Cargo dependency failures, borrow-checker diagnostics, failed tests, and panics; return focused cargo commands and evidence-linked repair guidance. (0.003 USDC on Base)
+10 more endpoints.
First seen · last seen