Debug Run Transcript Maintainer
Use when analyzing one Rudder agent run or recent run batches: run IDs, partial IDs, transcripts, run logs, execution traces, runtime failures, finalizer failures, stdout/stderr, run quality, or recent org run behavior.
Install
$ agentstack add skill-undertone0809-rudder-debug-run-transcript-maintainer ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-undertone0809-rudder-debug-run-transcript-maintainer) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Debug Run Transcript Maintainer
Overview
Diagnose one run or a recent run batch by reconstructing execution from run-intelligence, logs, and targeted DB evidence.
When to Use
Use this skill when:
- the user asks why an agent run failed
- the user names a run id, run prefix, org, runtime, or timeframe
- recent run quality or batch behavior needs analysis
- tool success and Rudder final status disagree
Do not use this skill when:
- claim a product fix is proven from transcript diagnosis alone
- skip run-intelligence and jump to raw DB guesses
Core Pattern
identify run/batch -> source priority -> failure classification -> evidence -> next routeQuick Reference
| Situation | Action | | --- | --- | | Single run | Resolve and reconstruct transcript | | Recent batch | Build bounded cohort and classify failures | | Finalizer failure suspected | Separate tool result from stream/persistence/UI status | | Fix required | Hand back to lifecycle verification path |
Implementation
- Identify run, org, runtime, or batch window.
- Use run-intelligence loader/API first, filesystem logs second, direct DB last.
- Classify root cause: model/runtime, tool call, parser, event stream, persistence, UI status, or continuity.
- Separate diagnosis evidence from fix proof.
- Report what happened, key evidence, and next route.
Reference files are part of this skill contract. Before executing high-risk actions or final judgments, load references/runbook.md for the detailed legacy workflow, examples, validation cases, and command-level guidance.
Use evals/ when the route needs that detail; keep the entrypoint thin.
Common Mistakes
| Mistake | Fix | | --- | --- | | Treating run debug as implementation | Debug first, then route fixes separately. | | Using DB rows as full transcript truth | Prefer run-intelligence/log source. | | Merging finalizer and tool failures | Diagnose them separately. | | Over-scanning recent runs | Bound the cohort and state filters. |
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: Undertone0809
- Source: Undertone0809/rudder
- License: Apache-2.0
- Homepage: https://rudder.zeeland.studio
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.