research-call-tracker
Log, update, and close published research calls and positions in shared-brain ledgers (Operations/Research/Call Tracker.md and Trade Journal.md), and compute the measured win rate. Use whenever a research note, market call, thesis, or trade idea is published or shipped; when a position opens, closes, or hits a stop/kill condition; or when asked "what's our win rate", "log this call", "update the tracker", or "close the position".
- License 许可
- MIT
- 最近更新
- 2026-11-10
- Stars
- ★ 13
- Forks
- 4
01 前言 Front Matter
- name
- research-call-tracker
- description
- Log, update, and close published research calls and positions in shared-brain ledgers (Operations/Research/Call Tracker.md and Trade Journal.md), and compute the measured win rate. Use whenever a research note, market call, thesis, or trade idea is published or shipped; when a position opens, closes, or hits a stop/kill condition; or when asked "what's our win rate", "log this call", "update the tracker", or "close the position".
02 资源文件 3
- 参考文档.hivemind-skill-source.json1.1 KB
- 参考文档templates/Call Tracker.md819 B
- 参考文档templates/Trade Journal.md1.0 KB
03 SKILL.md 正文
Research Call Tracker
Outcome accountability for research. A researcher who publishes calls without tracking them has no feedback loop between thesis quality and results — win rate becomes remembered instead of measured. This skill maintains two ledgers in the shared brain vault:
Operations/Research/Call Tracker.md— the research record: one row per published call.Operations/Research/Trade Journal.md— the position record: one row per actual position, with entry discipline.
If either ledger does not exist yet, seed it from the matching file in this skill's templates/ folder (copy verbatim, then set the Last updated date). Never invent a different table shape — future sessions and stats depend on the columns staying stable.
The publish-time rule (fail-closed)
A research call is not "published" until its Call Tracker row exists. Log the row at publish time, never retroactively — the entire value of the ledger is that conviction tier, entry zone, stop, and kill conditions were committed before the outcome was known. If a note ships through you and has no row, that is a blocker to surface, not a detail to skip.
Logging a new call
Append a row with: date (absolute, YYYY-MM-DD), asset, venue published to, conviction tier (High / Medium / Speculative — exactly one, always), entry zone, stop, target(s), status Watching or Open, thesis in one line, the sharpest kill condition, and a link to the adversarial verdict file (from the kill-my-thesis skill, if installed) or the note itself.
Journal entry: only if there is a real trade setup (entry + confirmation trigger + stop). Otherwise the call is research-only; skip is a valid outcome.
Updating and closing
Watching → Openwhen the confirmation trigger fires; the journal row must carry a live stop before it readsOpen.- Close with
Closed (Win)/Closed (Loss)/Closed (BE)plus outcome % in the tracker, and a one-sentence post-mortem in the journal: what the thesis got right, what it missed, disciplined or emotional exit. - A kill condition firing is a close signal, not a debate prompt. Surface it to the user the day it fires.
Stats
Recompute the tracker's Stats block whenever any call closes: total, closed count by outcome, win rate (wins ÷ closed, BE excluded from both sides), average return on closed. Only Closed rows count — open positions are hope, not results. Update the Last updated date.
Boundaries
- This is the decision record. Fill mechanics, paper trading, and live execution stay in the wallet/trading rails the user has configured.
- Never edit or delete historical rows to improve the record; corrections get a note in the row, not a rewrite.
- Ledger rows may name prices and levels but never keys, balances, or private network addresses.
04 同仓库更多 Skills
tailscale-generation-skill-authoring
This skill should be used when the user asks to "create a Tailscale generation skill", "add a local generation skill", "document a generation endpoint", "add an image generation provider", "add a TTS provider", "add video generation", "add 3D generation", or needs guidance on where HivemindOS Tailscale or local generation providers belong in the shared skills shelf.
localtts
This skill should be used when the user asks to "generate local TTS", "use local text to speech", "use remote TTS over Tailscale", "make speech audio", "add a TTS provider", "use Qwen3-TTS", or needs TTS endpoint setup through local or HivemindOS Tailscale routes.
karpathy-guidelines
Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.
comfyui-image-generation
This skill should be used when the user asks to "generate images", "generate images with ComfyUI", "use local ComfyUI", "use remote ComfyUI", "use Z-Image", "add an image generation provider", or needs image-generation endpoint setup through local or HivemindOS Tailscale routes.
wrangler
Cloudflare Workers CLI for deploying, developing, and managing Workers, KV, R2, D1, Vectorize, Hyperdrive, Workers AI, Containers, Queues, Workflows, Pipelines, and Secrets Store. Load before running wrangler commands to ensure correct syntax and best practices. Biases towards retrieval from Cloudflare docs over pre-trained knowledge.
workers-best-practices
Reviews and authors Cloudflare Workers code against production best practices. Load when writing new Workers, reviewing Worker code, configuring wrangler.jsonc, or checking for common Workers anti-patterns (streaming, floating promises, global state, secrets, bindings, observability). Biases towards retrieval from Cloudflare docs over pre-trained knowledge.