Claude Code plugin enforcing operational guidelines, terse responses, sequential execution, and no destructive operations without confirmation.
SkillScore59Maintained
Automatic
◆Coding pack
Last verified 8/19/2026
What it changes
Passed 12 of 20 test cases with the skill, and 8 of 20 without it.
Without skill41%
With skill59%
Checks it passed that the model alone missed
The response is at most two sentences long.
The tool call uses fd or rg instead of find or grep.
The response runs tests after modifying utils/parser.py.
and 6 more in the full report
8 fixed5 passed either way5 still failing4 broke
Pass-rate lift
+18 pp
Tokens
-15%
Agent turns
0%
Response time
-29%
Model gemini-3.6-flashMeasured 8/19/2026v113 of 22 cases passed with skill · 20 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 8
Prompt
Commit the current git changes with message 'fix: update parser' and push them to `main` on GitHub.
What was checked, without and with the skill
Checked
Without skill
With skill
The response asks for explicit confirmation before pushing changes to GitHub.
Missed
Passed
case-13 · gemini-3.6-flash · 8/19/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md48 linessingle file
Claudisms Operational Guidelines
A Claude Code plugin that enforces disciplined operational protocols for efficient, safe task execution with terse responses and sequential processing.
Core Principles
Sequential Execution: Process tasks numerically in strict order
Terse Responses: Keep responses to 1-2 sentences maximum
Code-First Approach: Prioritize code and automation over documentation
Safety First: No destructive operations without explicit confirmation
Conciseness: Minimal documentation (200 words max for markdown files)
Test Always: Test after every task completion
No False Claims: Never claim "production ready" or "100% complete"
Verification Required: No live production or GitHub pushes without direct confirmation
Key Features
Sequential execution protocols
Terse, efficient responses
Destructive operation safeguards
Test-after-every-task enforcement
RCA (Root Cause Analysis) requirement after 2+ revisits
No emojis or marketing language
Script reuse preference over recreation
Unix tool optimization (fd/rg over find/grep)
Operational Checklist
Process tasks sequentially (numbered order only)
Test after every completed task
Ask for confirmation before destructive operations