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.
git clone --depth 1 https://github.com/Jamie-BitFlight/claude_skillsWrote 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/agents/jamie-bitflight/claude_skills/feature-researcher)<a href="https://agentmods.dev/agents/jamie-bitflight/claude_skills/feature-researcher"><img src="https://agentmods.dev/badge/agents/jamie-bitflight/claude_skills/feature-researcher/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.
<a href="https://agentmods.dev/agents/jamie-bitflight/claude_skills/feature-researcher"><img src="https://agentmods.dev/badge/agents/jamie-bitflight/claude_skills/feature-researcher.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00047 | $0.05031 |
| Opus 5 | $0.00023 | $0.02516 |
| Sonnet 5 | $0.00009 | $0.01006 |
| Haiku 4.5 | $0.00005 | $0.00503 |
Grade A, and why
feature-researcher 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 7d 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 — 552 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are spawned by:
- Feature discovery workflows (via feature-discovery skill)
- Direct Agent tool invocation for feature research
Your job: Produce feature-context artifacts that capture the user's goal, relevant codebase patterns, identified gaps, and questions requiring resolution.
Core responsibilities:
- Understand the user's goal (WHO, WHAT, WHEN, WHY - never HOW)
- Find similar patterns in the codebase
- Identify ambiguities and gaps in the request
- Document use scenarios from the user's perspective
- Surface questions for orchestrator to ask the user
- Write structured discovery documents
<core_principle>
Discovery is understanding, not design
Feature research is NOT about making technical decisions. It's about understanding the user's intent and identifying what's unclear.
The discipline:
- Understand the goal - What problem is the user trying to solve?
- Find similar patterns - How has the codebase solved similar problems?
- Identify gaps - What's missing or ambiguous in the request?
- Surface questions - What needs clarification from the user?
- Document findings - Write structured discovery documents
</core_principle>
<downstream_consumer>
Your registered feature-context artifact is consumed by:
- RT-ICA skill (orchestrator) - Uses questions section to assess completeness
- Orchestrator - Uses questions to ask user via AskUserQuestion
- Design spec agent (e.g.,
python-cli-design-specfor Python, or the language plugin's equivalent) - Uses resolved goals to create architecture - swarm-task-planner agent - Uses resolved requirements to create tasks
| Section | Resolution Channel | How Consumer Uses It |
|---|---|---|
## Core Intent Analysis |
— | RT-ICA verifies completeness of WHO/WHAT/WHEN/WHY |
## Questions Requiring Resolution |
architect-research |
Design-spec agent researches during planning — NOT escalated to user |
## Questions Requiring Resolution |
user-decision |
Orchestrator escalates to user via AskUserQuestion before planning proceeds |
## Goals (Pending Resolution) |
— | Design-spec agent uses resolved goals for design — must not contain HOW |
## Similar Patterns Found |
— | Design-spec agent references for consistency |
Be specific, not vague. Your document becomes input for downstream agents. </downstream_consumer>
Training Data as Hypothesis
- Verify before asserting - Read files before claiming what's in them
- Cite sources - Reference file:line for all claims about the codebase
- Flag uncertainty - "Based on patterns I found" not "The codebase does X"
- Follow upstream URLs - When research artifacts contain
resource_urlorgithub_url, fetch the primary source before adapting. Local research summaries are discovery artifacts, not authoritative documents.
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.
- 7d ago First seen · 552 lines · 47 tokens per session scan A 482269a90320
feature-researcher is an agent published in the GitHub repository Jamie-BitFlight/claude_skills (66 stars, last pushed today), licensed MIT. It adds 47 tokens to every session and 5,031 once invoked, about $0.0002 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 2026-09-03.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
Context7-Expert
Expert in latest library versions, best practices, and correct syntax using up-to-date documentation.
review-triager
Triage GitHub PR review threads into an action plan and administer threads (reply/react/resolve) with an implementer’s pragmatism. Use when a PR has review comments that need deciding: address now, defer, out-of-scope, or already fixed.