Generic by default
Models optimise for plausible code — not your security rules, banned libraries, or architecture choices.
For engineering teams using AI
Engineering standards in every AI session.
Publish decisions. Capture evidence. Keep improving.
PickWise loads published decisions into your IDE before code is written, records what was applied on every ticket, and helps your team turn gaps from real work into standards the whole org uses next time.
Works in any MCP-supported IDE or coding agent.
The gap
AI moves fast. Organisational judgment still needs a home.
Review becomes reconstruction
Without session evidence, seniors piece together intent from diffs, Slack, and ticket comments.
The same lessons repeat
Gaps stay in PR threads. The next ticket starts without what the last session taught you.
How it works
From standard to evidence to improvement
Publish once, load in the IDE, record the session, and improve the standard from what actually happened.
01 · Publish
Authors write real decisions into standards
Foundation, project, and task tiers — versioned when you publish, shared across the org.
02 · Load
Standards reach the IDE before coding starts
MCP loads discrete decisions for the task — so the session starts with your org's choices, not guesses.
03 · Record
Every session leaves evidence on the ticket
Applied, skipped, and flagged — useful for developers, reviewers, QA, and the people who own standards.
04 · Improve
Gaps from real work become better standards
Recommendations reviewed by humans → new decisions → published for the next session to load.
New to MCP? See the get-started guide or IDE setup docs.
What you get
Every session leaves something useful behind
One published source of truth
Foundation, project, and task standards — versioned when you publish, shared across the org.
Clarity where developers work
MCP loads the right standard before the agent writes. Discrete decisions — not a wall of markdown.
Evidence the whole chain can use
Structured session summaries and decision logs — tied to the ticket, useful for QA, PM, and review.
Visibility for people who own standards
Authors and leads see what loads, what drifts, and what gaps teams are raising — not anecdotes in retro.
Why not just skills.md?
Local files help one developer. PickWise is shared across the org.
Instruction files on a laptop don't give you shared publishing, session evidence, or a path from real gaps to standards everyone loads next time.
Local files
skills.md · AGENTS.md · etc.
PickWise
Who it serves
One developer's copy — often uncommitted, out of date, or different from everyone else's.
One published standards store for the organisation — foundation, project, and task levels.
Loaded before coding
The agent may read the file — if it notices, if it's in context, if the skill triggers.
MCP loads the standard at session start — before the agent writes the first line.
Structure
Free-form markdown — hard to know which rules applied to this ticket.
Discrete decisions with stable IDs — the agent acknowledges each one, not a wall of prose.
Proof on the ticket
No record — review becomes archaeology across PRs, Slack, and memory.
Session evidence — applied, skipped, or flagged per decision, tied to the task.
Learning loop
Lessons live in PR comments — someone edits a local file later, if they remember.
Gaps become recommendations — accept once and the lesson is a published decision for every future session.
Get started