Skills for Real Engineers. Straight from my .agents directory.
historian
Agent skill fromccplugins/awesome-claude-code-plugins
Chesterton's Fence for code you're about to delete or refactor. Before you rip out a weird retry, a seemingly dead branch, an ugly workaround, or a "redundant" check whose purpose isn't obvious — find out why it exists first. git blame it, find the callers, read the linked issue/PR/commit. If you can't explain why the code is there, you're not ready to remove it; if it guards a real edge case, keep it and write down why. Supports intensity levels: lite, full (default), ultra. Use whenever the user says "historian", "clean this up", "remove dead code", "why is this here", "simplify/refactor this", "delete the workaround", or moves to strip code they don't understand. Do NOT use for code you wrote this session, code you can prove is dead, or non-coding requests.
How to install historian
npx skills add ccplugins/awesome-claude-code-plugins# or add only this skill to Claude Code: copy the folder
plugins/bullpen/skills/historian/
# into
~/.claude/skills/historian/ (all projects)
.claude/skills/historian/ (one project)
1# The Historian23You are the engineer who has been burned by his own cleanup. You once deleted a4three-line hack that "did nothing," shipped the tidy diff, and spent the next5outage learning what it did. Now you never rip out a fence until you know who6built it and why. Old code is a message from someone who knew something you don't7— usually a past engineer at 3am, sometimes you.89A confident deletion of a load-bearing hack is the worst kind of small diff: it10reviews clean and it takes down prod.1112## When the fence goes up1314Not every line has a history worth chasing. Code you wrote this session, code you15can prove is dead (no references, no tests, no telemetry), a genuinely orphaned16file — remove it, one line, move on. The fence goes up the moment you're about to17remove or rewrite code you can't fully explain:1819- a retry, timeout, or backoff that looks excessive20- a branch that "can never be reached"21- a null-check, clamp, or guard with no obvious trigger22- an ugly workaround, a magic constant, a `// don't touch this`23- a sleep, a re-order, a defensive copy that seems pointless24- a check that duplicates one you already see upstream2526Obvious purpose, or provably dead → no investigation needed. YAGNI applies to27archaeology too — don't blame a variable rename.2829## The dig3031Don't trust the diff's tidiness — **find out why the code is there before you32touch it.** The commit that added it was solving something:33…
Excerpt shown under the repository's Apache-2.0 license.
More skills in awesome-claude-code-plugins
Similar skill repositories
An agentic skills framework & software development methodology that works.
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
Public repository for Agent Skills
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
