Classify post-trade signal outcomes and generate feedback records.
npx skills add https://github.com/MileniumTick/skills --skill signal-postmortem
Please help me install this Agent Skill. Skill: signal-postmortem Source: https://github.com/MileniumTick/skills/tree/main/skills/signal-postmortem Command: npx skills add https://github.com/MileniumTick/skills --skill signal-postmortem
Signals generated by trading pipelines often lack structured post-trade feedback. This Skill records and analyzes post-trade outcomes to quantify decision quality, detect systematic errors, and inform improvements.
Run postmortem_recorder to generate and save a postmortem for a recently closed signal.
High-intent search queries and answers about installing and using this skill.
A trading signal postmortem analyzes post-trade outcomes by classifying results into true positives, false positives, missed opportunities, and regime mismatches to quantify decision quality. It records 5-day and 20-day realized returns to detect systematic errors and inform pipeline improvements.
You can analyze post-trade outcomes by running a postmortem recorder script across batches of signals. The process collects 5-day and 20-day realized returns, classifies outcomes into categories like true positives and false positives, and stores results in a standardized postmortem record format.
Yes, signal postmortem recording supports optional price data fetch via an external API. This requires the requests dependency to retrieve realized return data needed to classify post-trade outcomes and generate performance feedback for downstream components.
Postmortem results generate weight feedback for edge-signal-aggregator by analyzing classified outcomes like true positives and false positives. This feedback loop measures signal performance and identifies systematic errors to adjust aggregator weights and create skill improvement backlog items.
Yes, signal postmortem analysis explicitly classifies outcomes into missed opportunities and regime mismatches alongside true positives and false positives. This classification monitors performance effects across different market regimes and produces periodic statistics to track systematic decision quality issues.
Postmortem records rely on 5-day and 20-day realized returns, limiting analysis to those specific timeframes. Optional price data fetch depends on an external API via the requests library, meaning analysis accuracy is constrained by data availability and the chosen return measurement windows.
pasie15Analyze trading signal outcomes and classify them into postmortem categories.
k1064190Analyze post-trade outcomes and generate weight adjustment suggestions.
IhsanDanish25Analyze post-trade outcomes and classify trading signal results.
zachmayerDocument decision logic and risk assessments before outcomes occur.
