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 instructions/kkkano/finsight/agents-mdgit clone --depth 1 https://github.com/kkkano/FinSightWrote 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/instructions/kkkano/finsight/agents-md)<a href="https://agentmods.dev/instructions/kkkano/finsight/agents-md"><img src="https://agentmods.dev/badge/instructions/kkkano/finsight/agents-md.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 | $0.03255 | $0.03255 |
| Opus 5 | $0.01628 | $0.01628 |
| Sonnet 5 | $0.00651 | $0.00651 |
| Haiku 4.5 | $0.00326 | $0.00326 |
Grade A, and why
FinSight AGENTS.md 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 3d 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 — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
2026-03-08 增量架构说明（RAG 三层）
backend/rag/layering.py：三层 collection 解析与元数据归一化入口，负责memory / ws / kb、collection_kind、entity_scope、entity_key语义。backend/graph/nodes/execute_plan_stub.py：普通执行链的 Working Set / 长期 KB 写入与联合检索编排。backend/agents/deep_search_agent.py：DeepSearch 的ws:deepsearch:*working set 与kb:stock:*晋升/补查入口。backend/rag/hybrid_service.py：RAG 文档模型补充 layer / entity / ingest / promotion / fingerprint 字段，并支持多 collection 联检。backend/rag/observability_store.py、backend/rag/observability_runtime.py：把 layer / collection path / hit 元数据落进 observability，供 Inspector 和 DB Browser 回放。frontend/src/pages/RagInspectorPage.tsx：Inspector 首屏、run 详情、collection 浏览、DB Browser 直接展示三层语义。
2026-05-18 增量架构说明（单 Agent 质量合同）
backend/research/agent_quality_contract.py：单 Agent 质量合同入口，负责给 evidence 分配稳定agent_source:*、构造 claim、计算agent_quality指标。backend/agents/base_agent.py：标准research()流程会自动挂载evidence_quality.agent_quality；覆盖research()的 Agent 需要手动调用合同。backend/agents/fundamental_agent.py：原生输出 growth_quality / cash_flow_quality / eps_revision / balance_sheet_risk claims。backend/agents/news_agent.py：原生输出 catalyst_candidate / noise_or_secondary_signal / event_calendar claims。backend/agents/risk_agent.py：原生输出 risk_score / factor_exposure / stress_test claims。scripts/agent_quality_eval.py+tests/eval/agent_quality_cases.json：确定性 fixture eval gate，用于保存 before/after 结果并审计单 Agent 质量变化。
2026-05-20 增量架构说明（后端 Agent 能力诊断强化）
backend/graph/nodes/planner.py：Planner 输出agent_selection诊断，plan_ready/decision_note事件携带跳过原因、预算优先级、deepsearch 意图。PlanIR Schema 容错增强（PlannerSchemaShapeError+ retry prompt）。backend/graph/nodes/conversation_router.py：新增内幕/非公开信息安全边界（_query_requests_illicit_nonpublic_info），多轮对话历史 ticker 补全主题提示。backend/graph/nodes/chat_renderer.py：新闻引用兜底——当 plan 无新闻源时直接抓取文章，确保回复契约有可引用 URL。backend/research/debate.py：新增build_read_only_adjudications，输出 Bull/Bear/Judge 只读裁决产物。backend/research/agent_quality_contract.py：增强 low_source_quality / recovery action 诊断。
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.
- 3d ago First seen · 95 lines · 3,255 tokens per session scan A d081478e2849
FinSight AGENTS.md is an instructions file published in the GitHub repository kkkano/FinSight (53 stars, last pushed 1mo ago), licensed MIT. It adds 3,255 tokens to every session, about $0.0163 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 instructions, from other repositories
co-op AGENTS.md
Instructions for Afnanksalal/co-op, covering co-op agent guide, product contract, workspace map, required checks and engineering rules.
bitcoin-risk-brief AGENTS.md
AGENTS.md instructions for akarazhev/bitcoin-risk-brief, covering repository guidelines, project structure & module organization, build, test, and development commands, required project context and superpowers workflow.
otto AGENTS.md
AGENTS.md instructions for 0Smallcat0/otto, covering agents.md, purpose, non-negotiable boundary, working rules and codegraph.
equity-desk AGENTS.md
AGENTS.md instructions for maxrotemberg04-spec/equity-desk, covering equity-desk — agent instructions and hard rules.
JPOS-MCP CLAUDE.md
Instructions for mohisyed/JPOS-MCP, covering claude.md, project overview, build & run commands, install dependencies and run server (stdio transport, for claude desktop).
Aegisdesk AGENTS.md
Instructions for sitanshukr08/Aegisdesk, covering agent guide, start here, architecture boundary, working rules and task routing.