feat: add private AI tracing with AgentPond - #32
Open
marcusschiesser wants to merge 1 commit into
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
onInit()after the default Admin app existsnpx agentpond@0.5.2 initWhy AgentPond
AgentPond makes fwd2cal's AI event extraction observable without adding a separate hosted tracing backend. Its OpenInference spans can show which model handled an email, how many tokens it used, how long extraction took, and whether a retry or structured-output failure occurred. That is especially useful for diagnosing emails that do not become valid calendar events. Trace batches stay in project-controlled Firebase Storage and can be synced for local DuckDB analysis.
Prompt and completion bodies are redacted by default because forwarded emails can contain personal or confidential information.
Verification
npm run lintnpm run buildnpx tsc -p tsconfig.json --noEmitonInit()path with a synthetic Firebase configgpt-4o-minirequest and observed an OpenInferenceLLMspan with model and token metadata;input.valueandoutput.valuewere both__REDACTED__The repository's full test script was not run because it starts emulators, waits for an interactive Google OAuth login, and opens a browser. No deployment was performed.
Firebase Storage status
This repository does not currently configure Cloud Storage or Storage Rules in
firebase.json, so this PR does not invent a rule override or claim that production access is verified. Before deployment, the project needs a configured default bucket and maintainer verification that no client rule can read or writeagentpond/**. Live Firebase export,agentpond sync, and trace listing could not be verified without production bucket access.