Langfuse CLI Agent Audit — Report Index
Generated: 2026-05-22
CLI: langfuse (langfuse-cli 0.0.10)
Scope: Critical failure modes only
Scores
| Area |
Score |
| Failure mode average |
1.4/3 |
| Passing critical modes |
6/22 |
| Partial critical modes |
7/22 |
| Failing critical modes |
6/22 |
| Indeterminate critical modes |
3/22 |
| Readiness |
9/15 [C] |
Reports
| File |
Audience |
Purpose |
report-dev.md |
Langfuse CLI maintainers |
Fix list ordered by agent impact. |
report-agent-dev.md |
Developers integrating the CLI into agents |
Invocation guidance and workarounds. |
report-runtime.md |
Runtime agents |
Short operational brief before calling the CLI. |
report-issues.md |
Agent users and maintainers |
Concrete bugs and gaps observed during checks. |
Key Findings
- Parser and validation failures do not honor
--json; agents receive prose usage text for common errors.
- Exit status collapses validation, auth, command, and network failures into generic exit code 1.
- Destructive and mutating commands lack dry-run, idempotency, and machine-readable effect metadata.
- Missing-auth output does not provide
AUTH_REQUIRED, auth_methods, or setup guidance in structured form.
- Large-output protection is not declared; agents must manage
--limit and --fields themselves.
Source Artifacts
environment.mdreadiness.mdfindings.mdtrace.mdissues.md