OT
Outcome Tracker
🔵 Stable🕐 updated 2026-07-02
🔷 SkillSpec L3
⚙ ships an executable helper
pm-autopilot
Record the testable predictions inside a decision, then score them against reality later — so frameworks earn trust from outcomes, not vibes. Use when committing to a prioritisation, forecast, or plan (to log what it predicts), when asked to review what actually happened, or to compute how well-calibrated past RICE scores, forecasts, or bets have been. Produces a prediction record at decision time, and a calibration report with per-framework hit rates at review time.
What to give it
▸Mode — record (new decision), review (score due predictions), or calibrate (analyse the history)
▸Record mode: — the decision artifact (RICE table, forecast, launch plan, OKR set) and where records live (a `predictions/` folder in the Brain, or a JSON/markdown file in the repo)
▸Review mode: — the stored predictions plus current metric values for the due ones
▸Calibrate mode: — the prediction history (the calculator below reads it as JSON)
✅ The bar it holds itself to
Every skill in this library self-verifies — these are this skill's own quality checks, straight from its definition.
✓Every recorded prediction has all six fields — no "improve activation" without a metric, band, and date
✓Confidence was stated before the outcome was knowable, never backfilled
✓Review scored every due prediction, including the embarrassing ones — no silent skips
✓Unresolvables carry a reason, and the calibration report counts them separately from misses
✓Calibration numbers come from the calculator, not estimation
⚠️ What it refuses to do
Do not reinterpret a claim after the fact so it scores as a hit — the original wording is the contract
Do not record point estimates when the author thinks in ranges — bands are honest, points are theatre
Do not let a framework take credit for hits and blame "execution" for misses — score the prediction as made
Do not compute calibration on fewer than ~10 resolved predictions per framework — report "insufficient history" instead
Do not skip recording because the decision feels obvious — obvious bets that miss are the most valuable calibration data
Install
npx pm-claude-skills add --agent claude # or codex · cursor · gemini · hermes
# or one-line MCP (every skill, any client):
claude mcp add pm-skills -- npx -y pm-claude-skills-mcpRelated skills
🔌 Embed this skill
Drop this on your blog, docs, or site — it renders a "Run this skill" card:
<div data-pm-skill="outcome-tracker"></div>
<script src="https://mohitagw15856.github.io/pm-claude-skills/embed.js" async></script>💬 Discussion
Outcome Tracker is one of 625 open-source professional AI agent skills — all SkillSpec L3.
Try them all in the browser · ⭐ Star on GitHub · Browse the full catalog