More screenshots
Works with the AI tools you already use
🧠 Retrieval Trace Auditor
Audit real RAG evidence traces for missing expected sources, impossible citations, unsupported claims, stale evidence, unused context, and weak routing.
Secure checkout via Stripe
See it in action
You say
{"case_id":"policy-17","query_class":"temporal","route":"vector_only","retrieved_evidence":[{"id":"policy-2024","superseded_by":"policy-2026"}],"expected_evidence_ids":["policy-2026"],"cited_evidence_ids":["citation-ghost"],"claims":[{"claim_id":"claim-1","text":"The 2026 policy replaced the rule.","supporting_evidence_ids":["support-ghost"]}]}
Your agent does
Release gate: FAIL Critical: citation-ghost was cited but never retrieved. Critical: claim-1 declares support-ghost, which was never retrieved. High: policy-2026 was expected but absent; temporal query used vector_only. Boundary: semantic entailment and answer correctness remain unverified.
What you get
About this skill
Retrieval Trace Auditor inspects the evidence path behind a RAG or knowledge-agent answer. Supply structured JSONL cases with stable retrieved evidence IDs, expected evidence when known, citations, answer claims, declared support IDs, query class, and route. The included deterministic local helper detects expected evidence that was not retrieved, citations or claim support that never appeared in context, claims with no declared support, stale selected evidence, unused retrieved context, and temporal or relational questions routed through vector-only retrieval. It returns severity-ranked findings, per-case dispositions, a release gate, and a review memo. It deliberately does not claim to judge semantic entailment or factual correctness.
How to install
Drop the file into your AI Agent. Works with Claude, Cursor, ChatGPT, and 20+ more.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
30-day refund
Not a fit? Get your money back
More reviews for JustHandled Labs5.0(2)
Feedback buyers left on other skills by the same creator.
Trust & safety
Security scanned
Verified clean 15 days ago
- 30-day refund guarantee
- One-time purchase, yours forever
- Secure checkout via Stripe
Frequently Asked Questions
Browse More Skills
📘 README Generator
Generate a README from your actual repo, not a hallucinated guess. Scans your manifests (package.json, pyproject.toml, go.mod, Cargo.toml) for real install steps, scripts, and dependencies across Node, Python, Go, Rust, and Java. Documentation that matches the code instead of drifting from it.
🔎 Lead Enricher
Turn a company URL or name into a B2B lead profile: industry, size, tech stack, likely decision-makers, estimated email patterns, and buying signals like funding or hiring, each with a confidence score so guesses are labeled as guesses. One company at a time; for batch lists and CRM export, see Lead Enricher Pro.
✍️ Git Commit Writer
Generate Conventional Commit messages from your staged git diff: auto-detects type and scope from the changed files, flags breaking changes, and keeps work-in-progress out of the message. Clean, searchable history that plays nicely with automated changelog tools.
📊 Pipeline Reporter
Turn a messy CRM CSV export into a clean sales pipeline report: weighted forecast by stage, rep performance, and a close-date outlook, with column headers auto-mapped (ARR to Amount, AE to Owner) so inconsistent exports still work. The single-report version; for 12-month forecasting, CRM sync, and PDF, see Pipeline Reporter Pro.