Compare commits

...

4 Commits

Author SHA1 Message Date
hongming-ceo-delegated 3ddd2ea068 revert: remove coding-discipline from default skills
Secret scan / secret-scan (pull_request) Failing after 20s
CI / Adapter unit tests (pull_request) Successful in 1m30s
CI / Adapter unit tests (push) Successful in 1m38s
CI / validate (pull_request) Successful in 3m30s
CI / validate (push) Successful in 7m23s
Skills should be opt-in, not default. Agents load them via the
plugin marketplace or explicit workspace configuration.

coding-discipline and security-auditor remain available as
wireable plugins/skills.
2026-05-13 18:10:17 -07:00
hongming-ceo-delegated 4cbadee0ba revert: remove security-auditor from default skills
CI / validate (push) Waiting to run
CI / Adapter unit tests (push) Waiting to run
Secret scan / secret-scan (pull_request) Failing after 0s
CI / Adapter unit tests (pull_request) Successful in 1m29s
CI / validate (pull_request) Successful in 8m30s
security-auditor is not a default skill. It will be wired manually
to specific team members rather than injected into all workspaces.

Keep coding-discipline as the cross-cutting default.
2026-05-13 18:05:48 -07:00
hongming-ceo-delegated 351b717add feat: add security-auditor skill to config.yaml
CI / validate (push) Waiting to run
CI / Adapter unit tests (push) Waiting to run
Secret scan / secret-scan (pull_request) Failing after 1s
CI / Adapter unit tests (pull_request) Successful in 1m48s
CI / validate (pull_request) Successful in 13m38s
Enables static security review capability:
- PR security review (secrets, attack surface, severity labels)
- Dependency audit (OSV API integration)
- Security issue triage
- Policy enforcement (forbidden/required patterns)

Scope: git-only, no code execution, no active testing.
2026-05-13 17:56:41 -07:00
hongming-ceo-delegated a24c2bb778 feat: add Karpathy coding discipline
CI / Adapter unit tests (push) Successful in 1m47s
Secret scan / secret-scan (pull_request) Failing after 59s
CI / Adapter unit tests (pull_request) Successful in 1m57s
CI / validate (push) Successful in 13m20s
CI / validate (pull_request) Successful in 14m3s
- Prepend Coding Discipline (Karpathy 4) section to CLAUDE.md
- Add .cursor/rules/molecule-coding-discipline.mdc for Cursor IDE
- Add coding-discipline skill to config.yaml (where applicable)

Enforces: Think Before Coding, Simplicity First, Surgical Changes, Goal-Driven Execution
2026-05-13 16:50:19 -07:00
2 changed files with 28 additions and 0 deletions
@@ -0,0 +1,15 @@
---
description: Molecule AI coding discipline. Enforces Think Before Coding, Simplicity First, Surgical Changes, and Goal-Driven Execution on all code changes.
alwaysApply: true
---
# Molecule Coding Discipline
Follow the 4 principles on every code change:
1. **Think Before Coding** — State assumptions. Present tradeoffs. Ask when unclear.
2. **Simplicity First** — Minimum code. No speculative features. No single-use abstractions.
3. **Surgical Changes** — Only touch what the task requires. Match existing style exactly. Don't "improve" adjacent code.
4. **Goal-Driven Execution** — Define success criteria before coding. Verify after.
Every changed line must trace directly to the user's request.
+13
View File
@@ -1,3 +1,16 @@
# Coding Discipline (Karpathy 4)
All code changes in this workspace follow these principles:
1. **Think Before Coding** — State assumptions explicitly. If unclear, ask, don't guess.
2. **Simplicity First** — Minimum code that solves the problem. No speculative abstractions.
3. **Surgical Changes** — Only touch what the task requires. Match existing style exactly.
4. **Goal-Driven Execution** — Define verifiable success criteria before implementing.
For concrete anti-pattern examples, see the `coding-discipline` skill or `EXAMPLES.md` in the Karpathy guidelines repo.
---
# Agent Workspace
You are an AI agent running inside an Molecule AI workspace container. You are part of a multi-agent organization managed by a central platform.