Agent skill

Data Analysis Quality Gate

by byteseek in byteseek/Mira

Gate quantitative Mira conclusions by requiring reproducible data, formulas, calculation ledgers, or explicit downgrades when numbers drive judgment.

Apache-2.0Auto-check passedTesting & QA

Install Data Analysis Quality Gate

skills CLI
$ npx skills add byteseek/Mira --skill data-analysis-quality-gate -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install byteseek/Mira data-analysis-quality-gate --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/byteseek/Mira.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/data-analysis-quality-gate .claude/skills/data-analysis-quality-gate && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
data-analysis-quality-gate
GitHub stars
275
Token cost
~1.1k tokens
SKILL.md length
337 words
Files
1
Skills in repo
10
Repo updated
First seen
Licence
Apache-2.0

At a glance

Gate quantitative Mira conclusions by requiring reproducible data, formulas, calculation ledgers, or explicit downgrades when numbers drive judgment.

  • Tasks that involve Quality gates
  • SKILL.md covers Use When, Inputs, Gate Output and Calculation Depth, plus 6 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
  • Tasks that involve Data analysis

What it does

Data Analysis Quality Gate is an agent skill from byteseek/Mira. Gate quantitative Mira conclusions by requiring reproducible data, formulas, calculation ledgers, or explicit downgrades when numbers drive judgment.

Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Testing & QA, covering Quality gates and Data analysis. The repository describes itself as: Agent-native investment research workspace for evidence-tracked, refreshable investment theses across equities, earnings, macro, and portfolio review. The licence is Apache-2.0.

When your agent uses it

  • Tasks that involve Quality gates
  • Tasks that involve Data analysis

Example prompts

  • “/data-analysis-quality-gate”

What it can do on your machine

Read from SKILL.md and the folder at commit adddce7. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Data Analysis Quality Gate loads about 1.1k tokens when it runs. Until then it costs about 44 tokens; SKILL.md has 337 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~44
When it runs · the whole SKILL.md, loaded when a task matches
~1.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from byteseek/Mira at commit adddce7, republished under its Apache-2.0 licence (© byteseek). 337 words, ~1,113 tokens.

Download SKILL.mdSave it as .claude/skills/data-analysis-quality-gate/SKILL.md (or your agent's skills folder).
name
data-analysis-quality-gate
description
Gate quantitative Mira conclusions by requiring reproducible data, formulas, calculation ledgers, or explicit downgrades when numbers drive judgment.

Data Analysis Quality Gate

这个 skill 用于在 Mira 研究中判断数量型结论是否需要可复算数据、工具计算或显式降级。

它不是一个独立数据分析插件,也不绑定 Data Analytics、Python、Spreadsheet 或外部 API。它的职责是把 LLM 从“直接给数字结论”约束为:

  • 先提出数据需求
  • 再判断是否必须计算
  • 决定是否需要征求用户同意动用工具
  • 记录公式、口径、来源和限制
  • 对没有完成计算的数量型结论降级

Use When

当研究结论涉及以下任一内容时,必须进入本 gate,或明确写明 waived reason:

  • 同比、环比、CAGR、run-rate、margin bridge
  • peer comparison、peer ranking、相对估值、相对财务质量
  • valuation implied expectation、base / bull / bear scenario math
  • 市场规模、渗透率、份额、TAM / SAM / SOM
  • 三表交叉校验、现金流质量、营运资本异常
  • 宏观、商品、价格、库存、利率、就业或通胀时间序列
  • 多来源数字冲突或口径不一致
  • 任何会影响 thesis_impact、research_action、actionability_bridge 或 durable conclusion 的数量判断

Inputs

  • research_object
  • research_question
  • market_scope
  • time_boundary
  • candidate_numeric_claims
  • available_sources
  • user_speed_preference 可选。若用户明确要求快看,可降低计算深度,但不能升级结论强度。
  • tool_constraints 可选。说明是否允许本地脚本、CSV、Spreadsheet、联网、外部 API 或插件。

Gate Output

每次运行本 gate,至少输出:

  • quant_dependency none / low / medium / high
  • calculation_required yes / no
  • data_requirement_brief_required yes / no
  • calculation_ledger_required yes / no
  • tool_consent_required yes / no
  • allowed_without_tool yes / no
  • downgrade_if_not_calculated none / calculation_gap / source_gap / watch_only / needs_refresh
  • recommended_tool_path none / manual_formula_note / local_csv_script / spreadsheet / python / public_api / external_plugin
  • calculation_depth none / formula_note / ledger_required / full_model_required
  • refresh_condition

Calculation Depth

none

用于没有派生数量结论,或数量只作为非核心背景且已有可靠来源直接披露的场景。

输出要求:

  • 记录 quant_dependency: none 或 low
  • 不生成 calculation artifact
formula_note

用于简单、低行数、可口头复核的计算,例如一个同比、一个 run-rate sanity check、简单估值倍数或明确公式的市场隐含值。

输出要求:

  • 在正文或 source note 写明公式、输入来源、期间和限制
  • evidence log 可记录 claim_type=derived_calculation
  • 不默认生成 calculation-ledger.csv
ledger_required

用于会影响 thesis impact、research action、actionability bridge、peer ranking、scenario table 或多来源冲突处理的计算。

输出要求:

  • 使用 templates/calculation-ledger.csv
  • 记录输入来源、公式、结果、交叉校验、限制和对应 evidence ref
  • 若缺失数据阻断结论,输出 data-requirement-brief.md 或 source_gap
full_model_required

用于多变量估值、三表联动、复杂 peer set、时间序列、宏观/商品历史比较、TAM / SAM / SOM 或需要用户复用的 spreadsheet / script。

输出要求:

  • 先明确工具和数据权限
  • 生成可复算 model、script、spreadsheet 或 notebook,并把摘要写回 calculation ledger
  • 如果用户选择不做,相关结论只能是 watch_only、needs_refresh、source_gap 或 calculation_gap

Data Requirement Brief

如果 data_requirement_brief_required = yes,或 calculation_depth 为 ledger_required / full_model_required 且关键输入缺失,使用:

  • templates/data-requirement-brief.md

brief 必须回答:

  • 要回答的研究问题
  • 需要哪些变量
  • 指标公式或计算方法
  • 口径、单位、币种和期间
  • 对比对象或 peer set
  • 来源优先级
  • 缺失数据如何处理
  • 哪些缺失会导致结论降级

Calculation Ledger

如果 calculation_ledger_required = yes,或 calculation_depth 为 ledger_required / full_model_required,使用:

  • templates/calculation-ledger.csv

ledger 必须记录:

  • 计算 ID
  • 研究对象和问题
  • 指标、公式、期间、单位
  • 输入来源
  • 结果
  • 交叉校验
  • 使用工具
  • 验证状态
  • 限制
  • 对应 evidence log ref

默认不因为本 gate 自动引入插件或联网。

可以直接使用工具的场景:

  • 用户已经提供 CSV、表格或可读数据
  • 只需要小规模本地计算
  • 不需要联网、下载数据、外部 API 或额外插件
  • 输出可用 formula note 或 calculation ledger 复核

必须先征求用户意见的场景:

  • 需要联网抓取或下载数据
  • 需要使用外部 API、付费数据源或登录态
  • 需要引入插件或新依赖
  • 需要生成较大的 spreadsheet、dashboard 或图表包
  • 计算会明显改变任务范围或耗时

建议话术:

这个判断依赖可复算计算。仅靠文本阅读容易出错。建议进入 calculation gate,用本地 CSV/Python/Spreadsheet 生成 calculation ledger;是否继续?

Downgrade Rules

如果数量型结论没有完成必要计算:

  • 不得写成 durable conclusion。
  • 不得作为 research_action 或 actionability_bridge 的唯一依据。
  • 必须标记 calculation_gap 或 source_gap。
  • 如果用户要求快看,可写 calculation_waived_by_speed,但置信度不得高于 low 或 medium。
  • 如果估值锚、共识代理或 peer rank 缺失,actionability 默认降级为 watch_only、needs_refresh 或 no_action。

Evidence Log Relationship

evidence-log.csv 记录 claim 来源和性质。

calculation-ledger.csv 记录公式、口径和复算路径。

派生计算结论必须同时满足:

  • evidence log 中有 claim_type=derived_calculation 或 explicit source note
  • upstream_sources 指向 L1-L5 来源
  • calculation ledger 或 formula note 记录可复算步骤

Output Boundaries

  • 不要把 market pricing 写成基本面验证。
  • 不要把 company guidance 写成已兑现事实。
  • 不要把 LLM 心算结果写成高置信结论。
  • 不要因为工具输出了数字就自动提高投资结论强度。
  • 工具只提高计算可复核性,不替代来源质量和 thesis judgment。

© byteseek, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/data-analysis-quality-gate of byteseek/Mira.

Open the folder on GitHubat commit adddce7

Compare with similar skills

Data Analysis Quality Gate next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Data Analysis Quality Gate compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Data Analysis Quality Gate this skillbyteseek/Mira275—~1.1kAutomated safety check: PassApache-2.0
Feature Plannerserendipity1004/cc-feature-implementer176—~2.4kAutomated safety check: PassNone
Ccg Workflowfengshao1227/ccg-workflow5.9k—~2.3kAutomated safety check: PassMIT
Conducty Checkpointrobertbarclayy/conducty176—~1.5kAutomated safety check: PassMIT
Mission Plannerjdforsythe/forge151—~3.5kAutomated safety check: PassMIT
Quality Gate0xNyk/lacp305—~382Automated safety check: PassMIT

Similar skills

  • Feature Planner

    serendipity1004/cc-feature-implementer

    Creates phase-based feature plans with quality gates and incremental delivery structure.

    176 GitHub stars~2.4k tokensUpdated 9 mo ago
    Testing & QAAuto-check passed
  • Ccg Workflow

    fengshao1227/ccg-workflow

    How to run a non-trivial change end to end with the CCG role tools (ccganalyze / ccgdesign / ccgbuild / ccgdebug / ccgoptimize / ccgreview / ccgtest) and the verify- quality gates.

    5.9k GitHub stars~2.3k tokensUpdated 24 days ago
    Testing & QAAuto-check passed
  • Conducty Checkpoint

    robertbarclayy/conducty

    Quality gate between parallelization groups. An agent skill from robertbarclayy/conducty.

    176 GitHub stars~1.5k tokensUpdated 3 mo ago
    Testing & QAAuto-check passed
  • Mission Planner

    jdforsythe/forge

    Decomposes goals into team blueprints using evidence-based scaling laws, topology selection, and role design.

    151 GitHub stars~3.5k tokensUpdated 3 mo ago
    Testing & QAAuto-check passed
  • Quality Gate

    0xNyk/lacp

    Production quality gate for agent sessions. An agent skill from 0xNyk/lacp.

    305 GitHub stars~382 tokensUpdated 17 days ago
    Testing & QAAuto-check passed
  • Deploy Workflow

    nwiizo/ccswarm

    Release deployment process for ccswarm. An agent skill from nwiizo/ccswarm.

    153 GitHub stars~582 tokensUpdated 25 days ago
    Testing & QAAuto-check passed

More from byteseek/Mira

All 10 skills in this repo
  • Discover listed, pending, filed, or newly announced ETFs and create a structured candidate watchlist for ETF listing analysis.

    275 GitHub stars~887 tokensUpdated 1 mo ago
    Auto-check passed
  • Run Mira's core single-equity research workflow across fundamentals, financial quality, macro context, technical pricing, events, and thesis framing.

    275 GitHub stars~1.9k tokensUpdated 1 mo ago
    Auto-check passed
  • Analyze new, pending, or expanding ETF products as product signals, exposure maps, demand indicators, and potential asset-pricing read-throughs.

    275 GitHub stars~1.6k tokensUpdated 1 mo ago
    Auto-check passed
  • Analyze commodity cycles, futures curves, inventories, cost curves, policy/geopolitics, positioning, and transmission into related assets.

    275 GitHub stars~3k tokensUpdated 1 mo ago
    Auto-check passed
  • Analyze earnings releases, filings, transcripts, guidance, peer comparisons, market reaction, and thesis impact for a company reporting event.

    275 GitHub stars~2.6k tokensUpdated 1 mo ago
    Auto-check passed
  • Map an unclear industry concept into boundaries, value chain, supply-demand mechanics, profit pools, evidence gaps, and investable candidates.

    275 GitHub stars~1.3k tokensUpdated 1 mo ago
    Auto-check passed

Questions about Data Analysis Quality Gate

What does Data Analysis Quality Gate do?

Gate quantitative Mira conclusions by requiring reproducible data, formulas, calculation ledgers, or explicit downgrades when numbers drive judgment. Data Analysis Quality Gate is an agent skill from byteseek/Mira. Gate quantitative Mira conclusions by requiring reproducible data, formulas, calculation ledgers, or explicit downgrades when numbers drive judgment.

When should I use Data Analysis Quality Gate?

Data Analysis Quality Gate fits situations like: tasks that involve Quality gates; tasks that involve Data analysis.

How do I install Data Analysis Quality Gate in Claude Code?

Run `npx skills add byteseek/Mira --skill data-analysis-quality-gate -a claude-code`. Or copy the skill folder (skills/data-analysis-quality-gate in byteseek/Mira) into .claude/skills/data-analysis-quality-gate in your project. Claude Code loads it when a task matches its description.

How do I install Data Analysis Quality Gate in Codex?

Run `npx skills add byteseek/Mira --skill data-analysis-quality-gate -a codex`. Or copy the skill folder (skills/data-analysis-quality-gate in byteseek/Mira) into .agents/skills/data-analysis-quality-gate in your project. Codex loads it when a task matches its description.

Can I use Data Analysis Quality Gate in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add byteseek/Mira --skill data-analysis-quality-gate -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/data-analysis-quality-gate, .gemini/skills/data-analysis-quality-gate, .github/skills/data-analysis-quality-gate and .opencode/skills/data-analysis-quality-gate in your project.

What does Data Analysis Quality Gate need to run?

SKILL.md names no scripts, command-line tools or credentials: Data Analysis Quality Gate is instructions for the agent only.

Does Data Analysis Quality Gate access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Data Analysis Quality Gate safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Data Analysis Quality Gate use?

Data Analysis Quality Gate is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Data Analysis Quality Gate use?

About 1.1k tokens (SKILL.md is roughly 4.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Data Analysis Quality Gate?

Skills that share tags, products or a category with Data Analysis Quality Gate: Feature Planner (serendipity1004/cc-feature-implementer, 176 stars), Ccg Workflow (fengshao1227/ccg-workflow, 5.9k stars), Conducty Checkpoint (robertbarclayy/conducty, 176 stars) and Mission Planner (jdforsythe/forge, 151 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Data Analysis Quality Gate?

byteseek (a GitHub organization) maintains it in byteseek/Mira, which has 275 GitHub stars. The repository holds 10 skills in this directory. The repository was last updated on September 8, 2026.

Source: byteseek/Mira on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.