feat(observe): local dashboard — view real answers + run eval packs#2201
Open
Mikecranesync wants to merge 1 commit into
Open
feat(observe): local dashboard — view real answers + run eval packs#2201Mikecranesync wants to merge 1 commit into
Mikecranesync wants to merge 1 commit into
Conversation
…viewing & control) The original ask that kicked off this work stream: one local screen to see what MIRA does and run the test packs. Single file, stdlib http.server, loopback-only, no auth, no engine needed. - Evals tab: a button per eval pack → POST /api/run → run_eval.run() → live scorecard (status/asset/retr/cite/pts/warnings); lists past reports, click to reopen. - History tab: reads the newest traces-*.csv from the Langfuse export (tools/langfuse_export.py output) — browse the 3,725 real production answers, grounded% bar, filter grounded/ungrounded, search questions. - Reads simlab/observe/reports/*.json + the export CSV; writes nothing beyond what run_eval already writes. Binds 127.0.0.1 only. Verified live: all endpoints 200; /api/history reports 3725 answers / 2734 grounded; POST /api/run conveyor_demo → 7 pass / 2 fail; headless screenshot renders. ruff clean. Run: python -m simlab.observe.dashboard (→ http://127.0.0.1:8770) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01CS9fxC3gdSUJDJqHw1uMiu
Open
5 tasks
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Why
The original request that started this whole stream was "make the observe layer a dashboard (viewing + control)." This delivers it — and now there's real data to show (the 3,725 answers exported in #2172 + the eval packs).
What
simlab/observe/dashboard.py— single file, stdlibhttp.server, loopback-only, no auth, no engine needed.run_eval.run()→ live scorecard; past reports list, click to reopen.traces-*.csvfrom the Langfuse export — browse the 3,725 real production answers, grounded% bar, filter grounded/ungrounded, search.reports/*.json+ the export CSV; writes nothing beyond whatrun_evalalready does. Binds127.0.0.1only.Verification (live)
/api/history→ 3725 answers / 2734 grounded (73%);POST /api/run conveyor_demo→ 7 pass / 2 fail; headless screenshot renders;ruffclean.python -m simlab.observe.dashboard→ http://127.0.0.1:8770Notes
simlab/observe+run_eval); basefeat/observability-eval-layer. Retarget tomainafter feat(observe): production-grade observability + evaluation layer (phases 0-3) #2154 merges.MIRA_EXPORT_DIR→~/langfuse-export→tools/langfuse-export.🤖 Generated with Claude Code