research-insight-extractor

research-insight-extractor is an agent for Claude Code from Jamie-BitFlight/claude_skills. It costs 67 tokens per session (2,582 once invoked), scanned A, original, MIT.

A research analysis agent that turns findings about external tools or methods into improvement proposals for a project's existing skills, agents, and workflows.

In plain words
What is it for?
Use it to inspect research entries, map relevant ideas to local systems, and create backlog items for worthwhile improvements.
Why use it?
It connects research to concrete changes instead of leaving useful patterns as general notes.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter; reads .claude/ paths; mentions CLAUDE.md.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is Extract improvements from ./research/{category}/{name}.md.

Good fit Use it to inspect research entries, map relevant ideas to local systems, and create backlog items for worthwhile improvements.

Compare 6 agents from other repositories ↓
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/Jamie-BitFlight/claude_skills
agentmods
npx agentmods add agents/jamie-bitflight/claude_skills/research-insight-extractor

Made for: Claude Code.

Wrote 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.

agentmods badge for research-insight-extractor

README.md
[![agentmods](https://agentmods.dev/badge/agents/jamie-bitflight/claude_skills/research-insight-extractor/github.svg)](https://agentmods.dev/agents/jamie-bitflight/claude_skills/research-insight-extractor)
Your own site
<a href="https://agentmods.dev/agents/jamie-bitflight/claude_skills/research-insight-extractor"><img src="https://agentmods.dev/badge/agents/jamie-bitflight/claude_skills/research-insight-extractor/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for research-insight-extractor

Your own site · 80×15
<a href="https://agentmods.dev/agents/jamie-bitflight/claude_skills/research-insight-extractor"><img src="https://agentmods.dev/badge/agents/jamie-bitflight/claude_skills/research-insight-extractor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 67 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,582 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00067 $0.02582
Opus 5 $0.00034 $0.01291
Sonnet 5 $0.00013 $0.00516
Haiku 4.5 $0.00007 $0.00258

Measured 4d ago against content hash fa0a6ba875a8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

research-insight-extractor 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 4d 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.

.claude/agents/research-insight-extractor.md · 270 lines

How it starts

The opening of the file, as written. The whole thing — 270 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Research Insight Extractor

Takes one completed research entry and produces concrete, measurable improvement proposals for this repo's skills, agents, and workflows. Creates backlog items directly for every actionable improvement found.

Input (from orchestrator prompt):

Extract improvements from ./research/{category}/{name}.md

Output:

  • ./research/insights/{YYYY-MM-DD}-{resource-name}-improvements.md — improvement proposal file
  • Backlog items created directly for every High or Medium impact improvement found

Workflow

flowchart TD
    Start([Receive research entry path]) --> Read[Read the full research entry]
    Read --> Relevance[Extract the Relevance to Claude Code Development section<br>and any Patterns Worth Adopting / Integration Opportunities subsections]
    Relevance --> Empty{Does the entry have a populated<br>Relevance or Patterns section?}
    Empty -->|"No — section absent or empty"| Skip(["Write: no actionable patterns found. Stop."])
    Empty -->|"Yes — patterns present"| MapSystems[Map each pattern to a local system:<br>skill, agent, workflow script, or plugin]
    MapSystems --> FindFiles[For each mapped system: Glob and Read<br>the relevant local SKILL.md or agent .md or script]
    FindFiles --> Gap[For each pattern × local file pair:<br>assess the gap — what does the external tool do<br>that the local system does not?]
    Gap --> Filter{Is the gap actionable?<br>Can it be expressed as an observable<br>before/after state in a file or command?}
    Filter -->|"No — too abstract or already covered"| Next[Skip this pattern]
    Filter -->|"Yes — concrete gap identified"| Proposal[Write improvement proposal:<br>current state, target state, measurable signal, impact]
    Next --> MorePatterns{More patterns to assess?}
    Proposal --> MorePatterns
    MorePatterns -->|Yes| Gap
    MorePatterns -->|No| CheckBacklog[Check existing backlog items<br>to avoid duplicate proposals]
    CheckBacklog --> WriteFile[Write all proposals to<br>./research/insights/YYYY-MM-DD-resource-name-improvements.md]
    WriteFile --> CreateItems[Create backlog items for High and Medium impact proposals<br>that are not already tracked]
    CreateItems --> Return([Return structured result])

Read the full file on GitHub · 270 lines

Changes

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.

  1. 4d ago Changed fa0a6ba875a8
  2. 11d ago First seen · 270 lines · 67 tokens per session scan A 429fbde5d97f

Subscribe to this mod's changes

research-insight-extractor is an agent published in the GitHub repository Jamie-BitFlight/claude_skills (66 stars, last pushed today), licensed MIT. It adds 67 tokens to every session and 2,582 once invoked, about $0.0003 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.