Score breakdown
Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.
Why it matters
Useful for developers building AI agents (any Python agent framework) who need durable, observable traces per user-visible task and the ability to replay + diff failing runs locally: promptetheus is the MIT-licensed Python SDK that gives you one trace per agent task via decorators (`@pt.observe`, `@pt.tool`, `@pt.traced`), typed events for tool calls / browser actions / LLM calls / retrieval / sco
Who should use it
Who should skip it
Skip obro79/promptetheus if the source repository or demo is inactive, unmaintained, or no longer matches the description shown here.
About this signal
obro79/promptetheus is tracked by RepoRadar as a debugging infrastructure for ai in the MIT Python SDK + local replay CLI + hosted MCP e section. It was first seen on 2026-06-25 and last updated on 2026-06-25. The current verdict is 'try now' with a Silver tier and easy setup difficulty. Across RepoRadar's eight signals, obro79/promptetheus is strongest on workflow potential (9.3) and setup ease (8.8) and weakest on momentum (6.0) — a profile worth weighing against your own priorities. This page summarizes the evidence RepoRadar has captured from captured source metadata. The score, tier, risk label, and verdict on this page are never influenced by sponsorship, ads, or tips — they reflect only the usefulness, popularity, novelty, momentum, maturity, and evidence signals described in the RepoRadar methodology.
How this item is evaluated
RepoRadar assigned obro79/promptetheus a composite score of 7.8 out of 10, placing it in the Silver tier. This score combines weighted sub-signals: usefulness (35%), novelty (18%), momentum (14%), maturity (10%), open-source/build quality (7%), evidence quality (6%), workflow potential (6%), and setup ease (4%). Popularity is tracked separately at 29.0 and never affects the composite score or tier. The risk label of 'low' reflects inherent user-impacting hazards, not generic novelty. Items with no risk flag may still require normal code review before production use.
Risk explanation
**29 stars, 1 fork, 0 subscribers, early-stage.** promptetheus is at 29 stars with 1 fork and 0 subscribers, last push 2026-06-21 — moving fast but early-stage. Adopters in production should pin a specific PyPI version (e.g. `promptetheus==X.Y.Z`) and monitor the changelog for breaking changes to the typed event schema or the on-disk spool format; users who need a more battle-tested agent-observability layer should evaluate OpenTelemetry-native solutions first and consider promptetheus for new deployments where the decorator-based + durable-spool + hosted-MCP combination is the value proposition; **Hosted MCP evidence surface is read-only by design, not by accident.** The maintainer is explicit that the hosted MCP config snippets are read-only and scoped to a workspace + Supabase project — this is a deliberate security boundary so an agent reviewing its own failing runs cannot write back to telemetry storage. Adopters who need a writable MCP surface for agent-driven incident annotations must build that themselves on top of a separate write-enabled endpoint (do not promote the read-only MCP to a write surface without a separate audit trail); **`promptetheus spool purge` is a hard-delete operation.** The local spool is durable by default (events spool when HTTP delivery fails), so a long-running deployment accumulates spool files until purged. The maintainer documents `promptetheus spool purge` as a hard-delete operation — only run after confirming the data is no longer needed (e.g. after the build is green and the failure window has been reviewed).