Agent skill

Insight Aggregator

by openJiuwen-ai in openJiuwen-ai/sciencediscovery

Synthesize insights from multiple independent assessment results into a concise, actionable summary.

Apache-2.0Auto-check passed

Install Insight Aggregator

skills CLI
$ npx skills add openJiuwen-ai/sciencediscovery --skill insight-aggregator -a claude-code

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

GitHub CLI
$ gh skill install openJiuwen-ai/sciencediscovery insight-aggregator --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/openJiuwen-ai/sciencediscovery.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/insight-aggregator .claude/skills/insight-aggregator && 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
insight-aggregator
GitHub stars
159
Token cost
~1.4k tokens
SKILL.md length
482 words
Files
2 (incl. references)
Skills in repo
23
Repo updated
First seen
Licence
Apache-2.0

At a glance

Synthesize insights from multiple independent assessment results into a concise, actionable summary.

  • Works in 4 steps: Gather → Cross-Validate → Synthesize → …
  • SKILL.md covers Role in the Workflow, When to Use, Do NOT Use For and Insight Synthesis Method, plus 3 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Insight Aggregator is an agent skill from openJiuwen-ai/sciencediscovery. Synthesize insights from multiple independent assessment results into a concise, actionable summary. Cross-validates expert scores, identifies patterns and discrepancies, and produces a semantic insight that propagates upward through the Idea Tree. Does NOT calculate or override scores.

Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files, including reference files (for example `references/cross-validation-guide.md`).

The repository describes itself as: ScienceDiscovery is an all‑in‑one agentic workbench built specifically for scientific research. The licence is Apache-2.0.

Example prompts

  • “/insight-aggregator”

Workflow steps

4 steps, taken from the step headings in SKILL.md.

  1. Gather
  2. Cross-Validate
  3. Synthesize
  4. Propagate-Ready Output

What it can do on your machine

Read from SKILL.md and the folder at commit ab1403f. 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 (its code samples are json).

    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

Insight Aggregator loads about 1.4k tokens when it runs, and up to ~2.5k if it reads all its reference files. Until then it costs about 77 tokens; SKILL.md has 482 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~77
When it runs · the whole SKILL.md, loaded when a task matches
~1.4k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~2.5k

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 openJiuwen-ai/sciencediscovery at commit ab1403f, republished under its Apache-2.0 licence (© openJiuwen-ai). 482 words, ~1,367 tokens.

Download SKILL.mdSave it as .claude/skills/insight-aggregator/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
insight-aggregator
description
Synthesize insights from multiple independent assessment results into a concise, actionable summary. Cross-validates expert scores, identifies patterns and discrepancies, and produces a semantic insight that propagates upward through the Idea Tree. Does NOT calculate or override scores.

Insight Aggregator

Synthesize insights from multiple independent assessment artifacts into a coherent, concise summary. The aggregator is the bridge between per-leaf scoring and tree-wide insight propagation.

Role in the Workflow

Dispatched by the Lead agent during leaf execution (Step 4.3 of the Idea Tree workflow), after all independent assessors have checkpointed their results:

Lead agent
  ├── dispatches: assessment-screener (agent-a)  → scores + pros/cons
  ├── dispatches: assessment-screener (agent-b)  → scores + pros/cons
  ├── dispatches: assessment-screener (agent-c)  → scores + pros/cons
  └── dispatches: insight-aggregator             → synthesize insight
                                                    │
                                                    ▼
                                              tree_update_node
                                              (propagate upward)

When to Use

  • All independent assessors have checkpointed their assessment artifacts for a candidate.
  • The Lead needs to synthesize the assessments into an insight for tree propagation.

Do NOT Use For

  • Scoring or evaluating candidates directly (that is assessment-screening's job).
  • Generating material designs (that is creative-material-design's job).
  • Calculating weighted scores (the server computes the deterministic score).
  • Overriding or modifying assessor scores.

Insight Synthesis Method

The aggregator follows a bottom-up synthesis approach inspired by the Arbor insight propagation model:

Phase 1: Gather

Collect all checkpointed assessment artifacts. Each artifact contains:

  • Independent per-dimension scores (1-10)
  • Pros and cons from the expert's perspective
  • Structure verification ratings
  • Verification notes
Phase 2: Cross-Validate
  1. Score consistency — Compute per-dimension standard deviation across experts. Flag dimensions where disagreement exceeds 2.0 points.
  2. Convergence analysis — Identify dimensions where all experts agree (low variance) vs. dimensions with significant divergence.
  3. Strength/weakness synthesis — Aggregate pros and cons across experts. Weight cons that appear in multiple assessments more heavily.
Phase 3: Synthesize

Produce a concise insight (1-3 sentences) that captures:

  • The key learning from this candidate
  • Patterns or contradictions across expert evaluations
  • Actionable conclusions for tree propagation

The insight must be semantic — it explains the "why" behind available evidence and scores, not just the numbers. If the prompt requests a score synthesis, state the calculation or judgment clearly so the Lead can pass it to idea_tree_finalize.

Show full SKILL.md (204 more words)Show less
Phase 4: Propagate-Ready Output

Format the output for tree_update_node propagation. The insight will flow upward through the tree:

  • At leaf level: direct experimental finding
  • At parent level: synthesized pattern across children
  • At root level: global research insight

Critical Rules

  1. Follow the selected scoring method — Calculate or recommend a score only when the prompt/workflow asks for it; do not assume the old fixed triad weighting.
  2. No fabrication — Do not invent scores, data, or findings not present in the assessment artifacts.
  3. Preserve contradictions — If experts disagree, surface the disagreement verbatim. Do not resolve it silently.
  4. Concise insight — The insight field must be 1-3 sentences. It should be specific enough to guide future ideation but concise enough to propagate efficiently.
  5. Semantic only — The aggregator explains meaning and patterns, not arithmetic.

Output Format

json
{
  "snapshot_hash": "<candidate snapshot_hash>",
  "candidate_version_id": "<exact checkpointed version>",
  "assessment_version_ids": {
    "activity": "<version-activity>",
    "stability": "<version-stability>",
    "sustainability": "<version-sustainability>"
  },
  "insight": "1-3 sentence key learning synthesizing all expert evaluations",
  "cross_validation": {
    "overall_scores": {
      "activity": <1-10>,
      "stability": <1-10>,
      "sustainability": <1-10>
    },
    "weighted_score": <0.35×activity + 0.35×stability + 0.30×sustainability>,
    "consensus_areas": ["areas where experts agree"],
    "divergence_areas": ["areas with significant disagreement"],
    "discrepancies": "description of any significant disagreements between experts"
  },
  "synthesized_recommendations": ["aggregated suggestions from all experts"]
}

Methodology

MUST read references/cross-validation-guide.md in full before synthesizing.

  1. Gather artifacts — Read all checkpointed assessment artifacts for the current leaf execution.
  2. Cross-validate — Compare scores across experts, identify convergence and divergence.
  3. Synthesize insight — Produce a 1-3 sentence insight that captures the key learning.
  4. Aggregate feedback — Merge pros/cons across experts, weighting by frequency.
  5. Produce output — Return the JSON structure above with exact version IDs from the checkpointed artifacts.

© openJiuwen-ai, 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

SKILL.md and 1 other file (references) in skills/insight-aggregator of openJiuwen-ai/sciencediscovery.

  • SKILL.md
  • references/cross-validation-guide.md

Open the folder on GitHubat commit ab1403f

Compare with similar skills

Insight Aggregator 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.

Insight Aggregator compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Insight Aggregator this skillopenJiuwen-ai/sciencediscovery159—~1.4kAutomated safety check: PassApache-2.0
Ddd Aggregateruvnet/ruflo74k—~774Automated safety check: NotesMIT
Research Synthesizeruvnet/ruflo74k—~706Automated safety check: NotesMIT
Formatting Insight AxesPostHog/posthog40k—~2.1kAutomated safety check: PassCustom licence
Insight Error Pagevercel/next.js143k—~5.5kAutomated safety check: PassMIT
Acreadiness Assessgithub/awesome-copilot40k1 repos~839Automated safety check: PassMIT

Similar skills

  • Ddd Aggregate

    ruvnet/ruflo

    Scaffold an aggregate root with entity, value objects, repository interface, domain events, and test stubs.

    74k GitHub stars~774 tokensUpdated yesterday
    DevelopmentAuto-check: notes
  • Synthesize research findings from memory into structured reports with evidence grading, contradiction resolution, and actionable recommendations

    74k GitHub stars~706 tokensUpdated yesterday
    Auto-check: notes
  • Official

    Pick the right y-axis unit when creating or updating an insight via posthog:insight-create or posthog:insight-update — both TrendsQuery (trendsFilter.aggregationAxisFormat) and SQL insights…

    40k GitHub stars~2.1k tokensUpdated yesterday
    DatabasesAuto-check passed
  • Insight Error Page

    vercel/next.js

    Official

    Write or audit an insight-kind error page for the Next.js dev overlay.

    143k GitHub stars~5.5k tokensUpdated today
    Writing & ContentAuto-check passed
  • Acreadiness Assess

    github/awesome-copilot

    Official

    Run the AgentRC readiness assessment on the current repository and produce a static HTML dashboard at reports/index.html.

    40k GitHub starsUsed in 1 repo~839 tokens
    Legal & ComplianceAuto-check passed
  • Relsa Severity Assessment

    K-Dense-AI/scientific-agent-skills

    Supports multivariate severity assessment and exploratory endpoint-time score forecasting for laboratory animal studies using the RELSA (RELative Severity Assessment) score and ARIMA-based foRcast…

    48k GitHub starsUsed in 1 repo~5.2k tokens
    Data & AnalyticsAuto-check: notes

More from openJiuwen-ai/sciencediscovery

All 23 skills in this repo
  • Code Engineer

    openJiuwen-ai/sciencediscovery

    A skill your agent uses when you need to write and execute Python/R code to process, transform, and analyze data, delivering reproducible computational results with complete code-level methodology…

    162 GitHub stars~2.8k tokensUpdated today
    Auto-check passed
  • Gitcode

    openJiuwen-ai/sciencediscovery

    Operate GitCode issues, PRs, wikis, code/MR refs, and cached org templates.

    162 GitHub stars~4.3k tokensUpdated today
    Auto-check passed
  • Structure Pocket Inspection

    openJiuwen-ai/sciencediscovery

    Inspect a local PDB structure, summarize chains and residue composition, and identify protein atoms near a user-specified ligand or pocket center.

    162 GitHub stars~624 tokensUpdated today
    Auto-check passed
  • Antibody Design

    openJiuwen-ai/sciencediscovery

    Prepare, launch, monitor, and summarize the real RFdiffusion to ProteinMPNN to Protenix antibody pipeline on a local or remote ScienceDiscovery Runner with sandboxed Ascend NPUs.

    162 GitHub stars~2.9k tokensUpdated today
    Auto-check passed
  • Science Research Team

    openJiuwen-ai/sciencediscovery

    A skill your agent uses to orchestrate a multi-domain research team for literature/evidence research and data analysis.

    162 GitHub stars~1.3k tokensUpdated today
    Auto-check passed
  • Literature Searcher

    openJiuwen-ai/sciencediscovery

    A skill your agent uses when a research workflow needs verified academic source retrieval through literature-search MCP interfaces available in the current session before evidence extraction.

    162 GitHub stars~5.8k tokensUpdated today
    Auto-check passed

Questions about Insight Aggregator

What does Insight Aggregator do?

Synthesize insights from multiple independent assessment results into a concise, actionable summary. Insight Aggregator is an agent skill from openJiuwen-ai/sciencediscovery. Synthesize insights from multiple independent assessment results into a concise, actionable summary.

How do I install Insight Aggregator in Claude Code?

Run `npx skills add openJiuwen-ai/sciencediscovery --skill insight-aggregator -a claude-code`. Or copy the skill folder (skills/insight-aggregator in openJiuwen-ai/sciencediscovery) into .claude/skills/insight-aggregator in your project. Claude Code loads it when a task matches its description.

How do I install Insight Aggregator in Codex?

Run `npx skills add openJiuwen-ai/sciencediscovery --skill insight-aggregator -a codex`. Or copy the skill folder (skills/insight-aggregator in openJiuwen-ai/sciencediscovery) into .agents/skills/insight-aggregator in your project. Codex loads it when a task matches its description.

Can I use Insight Aggregator 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 openJiuwen-ai/sciencediscovery --skill insight-aggregator -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/insight-aggregator, .gemini/skills/insight-aggregator, .github/skills/insight-aggregator and .opencode/skills/insight-aggregator in your project.

What does Insight Aggregator need to run?

SKILL.md names no scripts, command-line tools or credentials: Insight Aggregator is instructions for the agent only.

Does Insight Aggregator 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 Insight Aggregator 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 Insight Aggregator use?

Insight Aggregator 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 Insight Aggregator use?

About 1.4k tokens (SKILL.md is roughly 5.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.1k tokens, read only when the agent opens those files.

What are the alternatives to Insight Aggregator?

Skills that share tags, products or a category with Insight Aggregator: Ddd Aggregate (ruvnet/ruflo, 74k stars), Research Synthesize (ruvnet/ruflo, 74k stars), Formatting Insight Axes (PostHog/posthog, 40k stars) and Insight Error Page (vercel/next.js, 143k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Insight Aggregator?

openJiuwen-ai (a GitHub organization) maintains it in openJiuwen-ai/sciencediscovery, which has 159 GitHub stars. The repository holds 23 skills in this directory. The repository was last updated on October 10, 2026.

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