Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/minihellboy/factorminer/research-ingestionnpx skills add minihellboy/factorminer --skill research-ingestiongit clone --depth 1 https://github.com/minihellboy/factorminerWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/minihellboy/factorminer/research-ingestion)<a href="https://agentmods.dev/skills/minihellboy/factorminer/research-ingestion"><img src="https://agentmods.dev/badge/skills/minihellboy/factorminer/research-ingestion.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00121 | $0.00802 |
| Opus 5 | $0.00060 | $0.00401 |
| Sonnet 5 | $0.00024 | $0.00160 |
| Haiku 4.5 | $0.00012 | $0.00080 |
Grade A, and why
research-ingestion scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Research Ingestion
This skill runs FactorMiner's Report-to-Memory Absorption (RMA) service — a scoped-down implementation of the RMA layer from XAlpha (arXiv:2607.08332): it screens external research fragments for OHLCV-representability, classifies the survivors into a broad mechanism family, and extracts reusable research-path hypothesis cues. It does not mine, generate, or backtest factors; it turns raw research text into structured input for factor-mining's generation prompts.
Why absorption instead of raw text
Feeding report text directly into a generation prompt lets ungrounded or OHLCV-infeasible claims (analyst EPS revisions, order-book microstructure, fundamentals) leak into hypothesis generation. RMA gates every fragment first, so only price/volume-representable mechanisms reach the mining loop.
The A/B/C pipeline
| Layer | Question | Output |
|---|---|---|
| A (eligibility) | Can this mechanism be observed, inferred, or proxied from daily OHLCV bars alone? | KEEP/DROP + reason |
| B (mechanism family) | Which broad mechanism bucket does it belong to? | One of factorminer.architecture.families.MECHANISM_FAMILIES |
| C (archetype) | What's the reusable research cue? | A ResearchArchetype record with research_paths |
DROPped fragments (fundamentals, analyst estimates, order-book state, news/sentiment, macro releases) are discarded — they are not representable under the daily OHLCV factor contract.
Workflow
1. Ingest a research note
factorminer ingest-research path/to/report_fragment.txt
Add --mock to run offline with the deterministic mock LLM provider (no API calls) — useful for smoke tests, never as a research result.
2. Read the classification
The command prints the KEEP/DROP verdict and reason. For a KEPT fragment it also prints the assigned mechanism family, fine-grained family, mechanism role, and research-path cues.
3. Hand off to mining
The resulting ResearchArchetype records are meant to be threaded into PromptContextBuilder.build(..., research_archetypes=[...]) so factor-mining's generation prompts carry the research-path text alongside memory and family context. Absorption itself never calls the mining loop — invoke factor-mining separately once you have archetypes worth exploring.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 51 lines · 121 tokens per session scan A ea6ef9a93fe7
research-ingestion is a skill published in the GitHub repository minihellboy/factorminer (107 stars, last pushed 6d ago), licensed MIT. It adds 121 tokens to every session and 802 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
sector-rotation
行业轮动分析——申万行业景气度评分、行业动量排名、产业链传导、估值/盈利/资金流多维比较框架.
strategy-pivot-designer
Detect backtest iteration stagnation and generate structurally different strategy pivot proposals when parameter tuning reaches a local optimum.
twitter-reader
Read Twitter/X for financial research using opencli (read-only). Use this skill whenever the user wants to read their Twitter feed, search for financial tweets, view bookmarks, look up user profiles, or gather market sentiment from Twitter/X. Triggers include: "check my feed", "search Twitter for", "show my…
chenhao-limit-up
Use when evaluating A-share limit-up (涨停板) setups through Chen Hao's sentiment and momentum lens: market emotion cycles, board strength, follow-through, and short-term aggressive momentum trading.
trading-risk-gate
Unified pre-trade safety gate: Ruin check (Law #1), ergodicity audit, and win-rate dominance validation. Absorbs: ergodicity-check, law-of-ruin, win-rate-dominance.
vectorbt
High-performance vectorized backtesting with parameter optimization, portfolio simulation, and rich performance metrics.