/SKILL.md
# LintGate Proactive professional cognition stack for LLM coding agents. Fires on every Write, Edit, and Bash via PostToolUse hooks. 125 MCP tools for on-demand analysis. ## First Steps 1. **Orient**: `build_theory_pack` — get the project's theory digest (~500-1500 tokens) 2. **Check**: `controlplane_run` — 7-channel supervision mesh (lint + tests + deps + git + behavior + structure + test_effectiveness) 3. **Reflect**: `constraint_check` — state your constraints before acting; `prediction_re
npx skillsauth add rohanvinaik/lintgate lintgateInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Proactive professional cognition stack for LLM coding agents. Fires on every Write, Edit, and Bash via PostToolUse hooks. 125 MCP tools for on-demand analysis.
build_theory_pack — get the project's theory digest (~500-1500 tokens)controlplane_run — 7-channel supervision mesh (lint + tests + deps + git + behavior + structure + test_effectiveness)constraint_check — state your constraints before acting; prediction_register — register falsifiable predictionsLintGate is not just a linter. It supervises the agent's reasoning strategy, not just its code output. The ControlPlane runs 6 independent channels in parallel — code quality, test health, dependency integrity, git hygiene, behavioral drift detection, and codebase structural analysis — with a coherence engine that reads the agreement and disagreement across channels as diagnostic signal. Theory extraction grounds findings in the project's own stated values. The Architecture of Inquiry closes the loop — behavioral discoveries flow back into the project's living semantic model. Habit Mode manages the context window itself — detecting sustained execution phases and producing structured compaction snapshots that turn context loss into context refinement.
bash setup.sh # venv, install, hook, MCP config, agent integration — one command
This detects which LLM agents are on the system and generates config files at the appropriate integration depth.
.claude/CLAUDE.md — Claude Code cognitive context: epistemic state, dispositions, guardrails, managed sections for living context. This is the deepest integration — behavioral discoveries flow back as patches..claude/rules/inquiry.md — Architecture of Inquiry protocoldocs/design.md — full architecture and design philosophyFor Claude Code, all three files work together. For other agents, AGENTS.md is the primary entry point and integrate.sh generates an appropriate cognitive context file.
tools
Use when work should span one or more detached tasks but still behave like one job with a single owner context. TaskFlow is the durable flow substrate under authoring layers like Lobster, ACPX, plugins, or plain code. Keep conditional logic in the caller; use TaskFlow for flow identity, child-task linkage, waiting state, revision-checked mutations, and user-facing emergence.
tools
# Lobster Lobster executes multi-step workflows with approval checkpoints. Use it when: - User wants a repeatable automation (triage, monitor, sync) - Actions need human approval before executing (send, post, delete) - Multiple tool calls should run as one deterministic operation ## When to use Lobster | User intent | Use Lobster? | | ------------------------------------------------------ | --------------------------
tools
# Lobster Lobster executes multi-step workflows with approval checkpoints. Use it when: - User wants a repeatable automation (triage, monitor, sync) - Actions need human approval before executing (send, post, delete) - Multiple tool calls should run as one deterministic operation ## When to use Lobster | User intent | Use Lobster? | | ------------------------------------------------------ | --------------------------
tools
A CLI tool for making authenticated requests to the X (Twitter) API. Use this skill when you need to post tweets, reply, quote, search, read posts, manage followers, send DMs, upload media, or interact with any X API v2 endpoint.