An AI coding agent skill designed to parse research reports, map critical metrics to an existing stock pool, and maintain deterministic tracking archives without manual data entry errors.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install report-tracker
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install report-tracker using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
Report Tracker is a robust financial analysis module optimized for Openclaw Skills. It automates the parsing of research reports, PDF summaries, and expert interview transcripts to maintain clear alignment with a predefined investment pool. By executing explicit validation and updating steps, it ensures all qualitative insights map directly back to target equities.
This skill eliminates human tracking errors by running a strict verification workflow. It scans research materials for core market forces like supply-demand gaps, timeline catalysts, downstream client movements, and technical route disputes. Crucially, the system works inside a deterministic sandbox, preventing unauthorized modifications or random internet queries from diluting carefully curated stock listings.
Ensure Openclaw Skills runtime is configured and your target workspace contains the required stock registry file.
Create the essential file hierarchy before running the agent process:
mkdir -p investment/archive
mkdir -p workspace/reports
mkdir -p references
touch investment/大开门股票池.md
touch investment/大开门股票池_研报跟踪存档.md
Populate your local tracking target registry investment/大开门股票池.md with explicit listings to prevent external search hallucination errors.
The skill records updates sequentially inside investment/大开门股票池_研报跟踪存档.md. Each data insertion conforms to this relational format:
| Attribute Field | Value Type / Constraint | Description |
|---|---|---|
| Timestamp | ISO-8601 String | Execution date-time marker |
| Report Source | String (Filename/URL) | Original provenance file reference |
| Linked Equities | String Array | Tickers matching the asset database |
| Connection Strength | Enum (Strong / Medium / Weak) | Impact category defining profitability adjustments |
| Key Discoveries | Text Block | Extracted data notes and catalysts |
| Valuation Shift | Numeric Delta | Absolute tracking changes (e.g., Old X Billion -> New Y Billion) |
investment/大开门股票池_研报跟踪存档.md (Main repository append-target)workspace/reports/研报关联分析_[YYYY-MM-DD].md (Consolidated multi-report brief)investment/archive/研报跟踪存档_[YYYY].md (Triggered automatically when the primary ledger passes 500 rows to archive entries older than 3 months)Loading
SkillOpt is a systematic optimization tool designed to treat agent skill files as trainable state, modifying them through measured rollouts, failure reflection, and rigorous validation gating.

An AI agent skill containing the core teachings, philosophies, and meditation techniques of the Diamond Sutra and the Sutra of Hui-neng.

An AI-powered historical research and book companion skill based on John U. Bacon's definitive account of the SS Edmund Fitzgerald shipwreck.

An AI-driven culinary science assistant based on J. Kenji López-Alt's 'The Food Lab' to help you master cooking through scientific principles.

An AI-driven SOP workflow designed to manage investment analysis, enforcing incremental updates on stock reports and maintaining strict research continuity.

An interactive knowledge and research engine designed to demystify the origins, mechanics, and controversies surrounding the Federal Reserve and fiat currency.








































