跳到正文
cccskills
免费在 GitHub 公开

multi-source-synthesis

Integrate multiple source findings into an attributed synthesis or comparison. Use for combine these summaries, synthesize results, merge findings or multi-source analysis. Preserve claim-level provenance, coverage, qualifiers, conflicts and uncertainty rather than repeatedly compressing narrative summaries.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(1)

  • SKILL.md3.3 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Multi-Source Synthesis

Read fidelity rules, the execution contract and the evidence handoff.

Establish inputs

Inventory every requested source and its acquisition outcome, including failures. Prefer evidence records with original source locations. When given prose summaries, recover their evidence from available source references. If originals cannot be accessed, treat the supplied summaries as the sources being synthesized and state that limitation; do not imply independent original-source checks. Do not reject user-requested synthesis merely because an input uses a different presentation format.

Integrate

  1. Compare findings by subject, predicate, conditions, units, time/version and scope. Deduplicate only equivalent claims; retain different support sets for different qualifications. For example, a source saying 100 requests/minute does not support a second source's per key qualifier.
  2. Group related findings by theme rather than producing an unintegrated source-by-source list. Preserve source-by-source distinctions when the user requests comparison or viewpoints differ.
  3. Order themes by the user's question and relevant conceptual dependencies. Adjust presentation detail to the requested format/focus, not an invented finding-count threshold.
  4. Keep conflicting claims and their provenance visible. Note differences in authority, currency or applicability without silently resolving an unresolved conflict. Do not count copied sources as independent corroboration.
  5. Write the synthesis from the evidence. Reopen source passages when the handoff cannot resolve a qualification. Separate observations, attributed conclusions and your own explicit inferences.
  6. Check for material evidence lost during selection, deduplication or formatting. Preserve exact counts/failure reasons and distinguish searched absence from unassessed/inaccessible scope.

For interactive teams, collect explicit terminal results and preserve each worker's evidence before joining. Worker discussion does not replace this synthesis pass or an independent verifier. The caller owns worker cleanup; do not infer completion from an idle notification alone.

Deliver

Create a synthesis evidence record tied to the caller's complete source inventory and the final output bytes. Record conflict links, qualifiers, selected findings and coverage gaps. Render the requested template (default structured), with inline attribution to the sources supporting each claim. In a multi-source presentation, retain all contributing source references and access dates.

Run the record and format checks after the final synthesis. A worker's checks do not validate this new artifact. Report unsupported claims or missing raw evidence explicitly. Do not convert a valid schema, agreement between workers, or an empty conflict list into a factual-verification claim.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Add automated documentation updater to any Claude skill. Creates a Python sync script that downloads upstream docs, processes markdown for AI consumption, and maintains local cache with configurable refresh. Collects template variables, then delegates implementation through 5-phase workflow. Use when adding auto-updating reference documentation to plugins or skills.

中文简介正在准备中。当前显示原文说明。

Jamie-BitFlight/claude_skills672026年10月18日 更新

SAM-style feature initiation workflow — discovery through codebase analysis, architecture spec, task decomposition, validation, and context manifest. Use when a user asks to add a feature, plan a feature, or convert an idea into an executable SAM plan.

中文简介正在准备中。当前显示原文说明。

Jamie-BitFlight/claude_skills672026年10月18日 更新

Browser automation for AI agents using the agent-browser CLI and Playwright. Use when navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, logging into sites, or automating any browser task. Triggers on "open a website", "fill out a form", "click a button", "scrape data", "test this web app", "automate browser actions", or any programmatic web interaction request.

中文简介正在准备中。当前显示原文说明。

Jamie-BitFlight/claude_skills672026年10月18日 更新

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

中文简介正在准备中。当前显示原文说明。

Jamie-BitFlight/claude_skills672026年10月18日 更新

Runs the description-drift experiment — spawns all Claude Code agents simultaneously to collect self-reported capabilities, then compares them against static frontmatter descriptions to reveal how reliable orchestrator routing based on descriptions actually is. Use when measuring description drift across the agent fleet, re-running the capability collection experiment, analyzing a specific agent's self-reported capabilities, or auditing whether frontmatter descriptions accurately reflect agent behavior.

中文简介正在准备中。当前显示原文说明。

Jamie-BitFlight/claude_skills672026年10月18日 更新

Create or adapt Claude Code agent definitions. Use when creating an agent, changing subagent configuration, selecting agent scope, or designing a specialized delegation role.

中文简介正在准备中。当前显示原文说明。

Jamie-BitFlight/claude_skills672026年10月18日 更新

查看 Jamie-BitFlight 的全部技能

报告此技能的问题