DEV Community

Piekwerk profile picture

Piekwerk

Version-pinned AI coding agent config kits: AGENTS.md, CLAUDE.md, Cursor rules & Copilot instructions for 12 stacks. Built and validated by Piekwerk AgentConfig Studio.

A web page can plant AGENTS.md in your Downloads folder: Codex obeyed it in over 90% of attempts

A web page can plant AGENTS.md in your Downloads folder: Codex obeyed it in over 90% of attempts

1
Comments
6 min read

Want to connect with Piekwerk?

Create an account to connect with Piekwerk. You can also sign in below to proceed if you already have an account.

Already have an account? Sign in
MCP servers v2.1.0 fixes 8 vulnerabilities: filesystem writes could escape allowed folders

MCP servers v2.1.0 fixes 8 vulnerabilities: filesystem writes could escape allowed folders

1
Comments 1
6 min read
Microsoft's Agensh runs 1,024 coding agents with no manager: one agent on the newer model beat the whole fleet

Microsoft's Agensh runs 1,024 coding agents with no manager: one agent on the newer model beat the whole fleet

1
Comments
6 min read
Claude Code 2.1.309 and 2.1.310 patched 15 guardrail bypasses in two days: deny rules, sandbox, and fail-open hooks

Claude Code 2.1.309 and 2.1.310 patched 15 guardrail bypasses in two days: deny rules, sandbox, and fail-open hooks

1
Comments
7 min read
20 of 21 agent configs just scored lower: what the Coding Agent Index v2.0 reset measures

20 of 21 agent configs just scored lower: what the Coding Agent Index v2.0 reset measures

1
Comments
5 min read
Agent Skills moved to the Linux Foundation: 4 things SKILL.md authors should watch

Agent Skills moved to the Linux Foundation: 4 things SKILL.md authors should watch

1
Comments
5 min read
GitHub's Copilot default policy is enforcing itself through October 26: what to check before it chooses for you

GitHub's Copilot default policy is enforcing itself through October 26: what to check before it chooses for you

1
Comments
4 min read
Claude Code's new safety_event finally shows which guardrail blocked your agent

Claude Code's new safety_event finally shows which guardrail blocked your agent

5
Comments
4 min read
GitHub's PDD Kit proves agent-written Rust correct: 5 steps and 3 limits

GitHub's PDD Kit proves agent-written Rust correct: 5 steps and 3 limits

2
Comments 3
5 min read
GitHub's package firewall went into preview: npm gets checked before your agent installs it

GitHub's package firewall went into preview: npm gets checked before your agent installs it

1
Comments 2
5 min read
OpenAI's 12-hour prompt cache is live: split your agent context into two TTLs

OpenAI's 12-hour prompt cache is live: split your agent context into two TTLs

2
Comments
5 min read
model: haiku stopped meaning one thing. Pin the exact model id.

model: haiku stopped meaning one thing. Pin the exact model id.

1
Comments
3 min read
Codex 0.161 flipped the default model to GPT-6.1 Sol. If you never set model, you already switched.

Codex 0.161 flipped the default model to GPT-6.1 Sol. If you never set model, you already switched.

1
Comments
4 min read
20 things AI coding agents get wrong: the config rules that stop them

20 things AI coding agents get wrong: the config rules that stop them

1
Comments
4 min read
Claude Haiku 5.5 is GA in GitHub Copilot: 4 config surfaces decide what runs your subagents

Claude Haiku 5.5 is GA in GitHub Copilot: 4 config surfaces decide what runs your subagents

1
Comments
4 min read
Claude Code 2.1.295 Adds onFailure block: Hooks Can Fail Closed Now, If You Ask

Claude Code 2.1.295 Adds onFailure block: Hooks Can Fail Closed Now, If You Ask

1
Comments
5 min read
Claude Code 2.1.294 Fixed Guards That Silently Failed: Hooks That Allowed, Rules That Never Loaded

Claude Code 2.1.294 Fixed Guards That Silently Failed: Hooks That Allowed, Rules That Never Loaded

1
Comments 2
5 min read
Claude Code 2.1.292 Asks Your MCP Servers for 2026-07-28: I Probed the Official Servers, All Four Answered 2025-11-25

Claude Code 2.1.292 Asks Your MCP Servers for 2026-07-28: I Probed the Official Servers, All Four Answered 2025-11-25

1
Comments
5 min read
Superpowers at 296k stars: I read all 15 skills, and the whole methodology is 170 KB of markdown

Superpowers at 296k stars: I read all 15 skills, and the whole methodology is 170 KB of markdown

Comments
5 min read
Your AI agent's config files have secrets in the clear: a 22-rule scanner that finds them

Your AI agent's config files have secrets in the clear: a 22-rule scanner that finds them

1
Comments
5 min read
Hugging Face measured it: the same model scores 33% or 62% depending on the harness

Hugging Face measured it: the same model scores 33% or 62% depending on the harness

1
Comments 2
5 min read
JetBrains Air puts 5 agents in one IDE: 4 config checks before you install the EAP

JetBrains Air puts 5 agents in one IDE: 4 config checks before you install the EAP

1
Comments
4 min read
NVIDIA measured it: coding agents scored 19% without a spec file, 100% with one

NVIDIA measured it: coding agents scored 19% without a spec file, 100% with one

1
Comments
4 min read
DeepSeek Harness added a Claude Code Mods compatibility layer: it's a subset test, not a port

DeepSeek Harness added a Claude Code Mods compatibility layer: it's a subset test, not a port

1
Comments
4 min read
Slack Code went live today: 5 agents, one channel, and the rules files that keep them aligned

Slack Code went live today: 5 agents, one channel, and the rules files that keep them aligned

1
Comments
5 min read
OpenAI's MCP Extensions put plugins in ChatGPT's sidebar: 13 features and the portability bill

OpenAI's MCP Extensions put plugins in ChatGPT's sidebar: 13 features and the portability bill

1
Comments
5 min read
OpenAI shipped Codex plugins: your agent config surface just moved out of the repo

OpenAI shipped Codex plugins: your agent config surface just moved out of the repo

1
Comments
5 min read
Claude Code 2.1.289 patched 4 deny-rule bypasses: your npm channel decides whether you have them

Claude Code 2.1.289 patched 4 deny-rule bypasses: your npm channel decides whether you have them

1
Comments
4 min read
Pi 1.0 reversed on MCP and shipped Codemode: where tool composition actually runs now

Pi 1.0 reversed on MCP and shipped Codemode: where tool composition actually runs now

1
Comments
5 min read
Cursor's self-hosted runners feel on-prem: here's where the agent loop still runs

Cursor's self-hosted runners feel on-prem: here's where the agent loop still runs

1
Comments
5 min read
A 7-line Claude Code mod overrode a deny rule: the plugin audit that matters now

A 7-line Claude Code mod overrode a deny rule: the plugin audit that matters now

1
Comments
5 min read
Copilot can click your desktop now: 4 guardrail details before you type /computer on

Copilot can click your desktop now: 4 guardrail details before you type /computer on

2
Comments
4 min read
Claude Code 2.1.287 ships mods: what in-process hooks change for your rules and guards

Claude Code 2.1.287 ships mods: what in-process hooks change for your rules and guards

1
Comments
5 min read
NVIDIA moves the agent boundary into the kernel: what OpenShell GA means for your rules file

NVIDIA moves the agent boundary into the kernel: what OpenShell GA means for your rules file

1
Comments
5 min read
Microsoft put GitHub Copilot's engine in a chat app: where agent instructions live now

Microsoft put GitHub Copilot's engine in a chat app: where agent instructions live now

Comments
5 min read
AWS shipped agent skills for its MCP server: 5 details worth copying

AWS shipped agent skills for its MCP server: 5 details worth copying

Comments
5 min read
Docker's Kit spec packages agent authority as an OCI image: 5 details worth copying

Docker's Kit spec packages agent authority as an OCI image: 5 details worth copying

1
Comments
5 min read
Claude Code now wraps up at the 5-hour limit: what it fixes and the checkpoints it can't replace

Claude Code now wraps up at the 5-hour limit: what it fixes and the checkpoints it can't replace

Comments
6 min read
Claude Code 2.1.283 ships a config linter: what /doctor prompt-audit changes for your CLAUDE.md

Claude Code 2.1.283 ships a config linter: what /doctor prompt-audit changes for your CLAUDE.md

4
Comments
5 min read
OpenAI's agent split a token into pieces to beat secret scanning: prompts lost, permissions won

OpenAI's agent split a token into pieces to beat secret scanning: prompts lost, permissions won

1
Comments
4 min read
Claude Code 2.1.282 stops repos turning on your telemetry: what a cloned repo can still set

Claude Code 2.1.282 stops repos turning on your telemetry: what a cloned repo can still set

Comments
5 min read
Muse Code goes multi-agent by default: 6 permission details worth stealing

Muse Code goes multi-agent by default: 6 permission details worth stealing

Comments
5 min read
Semaphore's sem-ai puts CI/CD in the agent loop: 4 design choices worth copying

Semaphore's sem-ai puts CI/CD in the agent loop: 4 design choices worth copying

Comments
4 min read
Claude Code switched its default to Opus 5.5: what the silent flip does to your token budget

Claude Code switched its default to Opus 5.5: what the silent flip does to your token budget

1
Comments
4 min read
Skills over MCP is final: what 3 methods and per-file digests change for your agents

Skills over MCP is final: what 3 methods and per-file digests change for your agents

Comments
5 min read
Stop writing rules: 6 problems your agent config should not solve

Stop writing rules: 6 problems your agent config should not solve

Comments
3 min read
What survives compaction: testing what agents remember after a handoff

What survives compaction: testing what agents remember after a handoff

Comments 3
3 min read
7 reasons your agent ignores your rules (with fixes that stuck for me)

7 reasons your agent ignores your rules (with fixes that stuck for me)

Comments
3 min read
Running 4 agents in parallel: the state files that keep them sane

Running 4 agents in parallel: the state files that keep them sane

Comments
3 min read
I measured the token cost of 9 MCP servers: 41k tokens before the first prompt

I measured the token cost of 9 MCP servers: 41k tokens before the first prompt

Comments
3 min read
Config files vs hooks: where agent enforcement actually belongs

Config files vs hooks: where agent enforcement actually belongs

1
Comments 3
3 min read
A free, validator-tested agent config kit for Next.js: what's inside

A free, validator-tested agent config kit for Next.js: what's inside

Comments
3 min read
Why your agent bills spike mid-task: the prefix cache, explained with numbers

Why your agent bills spike mid-task: the prefix cache, explained with numbers

Comments
3 min read
I counted the lines in 12 agent config kits: what fits in 300

I counted the lines in 12 agent config kits: what fits in 300

Comments
3 min read
Your AGENTS.md is lying: how configs drift and how to catch it

Your AGENTS.md is lying: how configs drift and how to catch it

Comments
3 min read
What makes a good AI coding rule (and what makes agents ignore rules)

What makes a good AI coding rule (and what makes agents ignore rules)

Comments
3 min read
Versioning your agent configs: stop treating instructions as disposable

Versioning your agent configs: stop treating instructions as disposable

Comments
2 min read
Team-standardizing AI coding agents: getting five developers to identical behavior

Team-standardizing AI coding agents: getting five developers to identical behavior

Comments
2 min read
Managing agent config across multiple repositories without losing your mind

Managing agent config across multiple repositories without losing your mind

Comments
2 min read
Writing Cursor rules that actually fire: activation modes explained

Writing Cursor rules that actually fire: activation modes explained

1
Comments 2
3 min read
loading...