Detect streamed tool-call anomalies without quantization claims #102
No reviewers
Labels
No labels
accessibility
bug
documentation
duplicate
enhancement
good first issue
help wanted
invalid
question
wontfix
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
nsaspy/llm-log!102
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "feat/58-tool-call-anomaly"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Tracks the first bounded detector family from #58, stacked on the verified #61 transport/policy stack.
What changed
wea+weather+weatherreconstructs asweather, not duplicated text;tool_arguments_invalid_jsonmodel-behavior evidence for complete reconstructed arguments that fail JSON decoding;tool_name_not_requesteddetector when the reconstructed function name is absent from the request's declared tool/function set;anomalies.jsonlthrough a dedicatedAnomalyActor;unknown; an explicitly grounded quantization value may annotate the observation but does not change detector ID or score;RED evidence
Initial test-only head
d9660780546156ad14391815192f5713d1318f7f: 134 tests ran; exactly six new tool-call contracts errored becausellm_log.tool_call_anomalydid not exist. Every inherited transport/routing/alert contract stayed green.Hardening test head
ef16bf098f35c8e0f437aba2f1ac5804bb5cf0d5: 136 tests ran; exactly two failures exposed real missing behavior:weaweatherweatherinstead ofweather;anomalies.jsonl.Everything else remained green.
GREEN evidence
528c6e5cc4494ffccbc55260f7b3819f3bc51725: Analytics and Quota exact-head workflows both passed, including delivered package build.a9ce4d44f9c0d234d8eafc110cc6984a419009de: Analytics run 84 and Quota run 89 passed, including full Python regression, adversarial analytics, delivered package build, cumulative-name normalization, live anomaly persistence, and the cross-domain regression proving transport-truncated malformed tool calls emit onlystream_protocol_errorand zero model anomalies.0b910c52121b0a37c6395961d0994013da8a5b99: Analytics run 89 and Quota run 94 both passed after Org-roam/Prolog knowledge promotion; Quota again built the delivered Python package.Boundary
This slice is observe-only model-behavior evidence. An anomaly is not automatically task failure, is not a transport error, and is not a bit-width detector. JSON Schema argument conformance, structured response-format violations, repetition/loop detection, output-length baselines, encoding corruption, self-correction signals, canary drift and any external anomaly notification transport remain separate #58 slices.
No merge is performed here.
View command line instructions
Checkout
From your project repository, check out a new branch and test the changes.Merge
Merge the changes and update on Forgejo.Warning: The "Autodetect manual merge" setting is not enabled for this repository, you will have to mark this pull request as manually merged afterwards.