skills.rest
lucasheriques
lucasheriquesCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

why

Investigate design rationale behind code by querying git history and MCP evidence sources in parallel.

Updated Sep 23, 2026
Search Tags:#mcp#design rationale#git history#code investigation#evidence synthesis#code archaeology#incident postmortem
One-click install
npx skills add https://github.com/lucasheriques/pstack --skill why-lucasheriques
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: why
Source: https://github.com/lucasheriques/pstack/tree/main/skills/why
Command: npx skills add https://github.com/lucasheriques/pstack --skill why-lucasheriques

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Code rarely explains why it exists. This Skill reconstructs the motivation and intent behind a piece of code by searching source control, issue trackers, docs, chat, observability, error tracking, and analytics sources, then returns a cited, confidence-weighted answer instead of a guess. ## Core Features & Use Cases - Parallel evidence investigation: Spawns one read-only investigator subagent per available evidence category (git/GitHub, Linear/Jira, Notion/Confluence, Slack, Datadog, Sentry, Databricks), each following a source-specific playbook. - Calibrated confidence output: Synthesizes findings into Direct, Supported, Inferred, Speculative, and Unknown tiers with verbatim citations, explicit gaps, and a Sources Consulted coverage map. - Use Case: Ask why a rate limit is clamped at 100, and the Skill traces the PR, linked ticket, Slack thread, and Datadog monitor to show the threshold came from an incident postmortem, citing each source. ## Quick Start Ask the AI why a specific function, threshold, or feature flag exists and request a cited investigation of its design rationale.

Frequently Asked Questions about why

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find out why a piece of code was written a certain way?▼

Anchor the target with git blame and git log to find commits and PR numbers, then search each available evidence source in parallel. The Skill queries source control, tickets, docs, chat, and telemetry MCPs and synthesizes a cited answer.

What is the difference between the how and why skills?▼

The how skill explains what code does and how it works at runtime. The why skill explains the forces that led to its shape, using historical evidence from commits, PRs, tickets, docs, and incidents.

Which MCP servers does the why skill support?▼

It maps available MCPs to seven categories: source control, issue trackers like Linear or Jira, docs like Notion or Confluence, chat like Slack, observability like Datadog, error tracking like Sentry, and warehouses like Databricks or Snowflake.

What happens if no MCP is available for an evidence category?▼

The category is recorded as an explicit gap in the Sources Consulted section rather than skipped silently. The final answer states which sources were not searchable and why, so coverage is transparent.

Can the why skill answer questions with no supporting evidence?▼

Yes, it reports an honest Unknown with specifics about what was searched and found empty. It never fills gaps with confident guesses, and speculative hypotheses are clearly labeled as such.

Related Skills

Explore More Software Engineering Skills→
latestaiagentslatestaiagents
5

git-history-detective

Investigate Git commit history with git blame, log, and bisect.

Official
Intermediate
rawhit-rrawhit-r

parallel-cross-service-fix

Dispatch one agent per microservice to apply isolated cross-service fixes in parallel.

Community
Advanced
rockcookiesrockcookies
1

how

Generates architectural explanations of how codebases work using parallel explorer subagents.

Community
Intermediate
japurcelljapurcell

commit-to-pr

Create a single conventional commit and GitHub PR from local changes.

Community
Advanced