Correction Intent Tagger

Detects when a user message corrects a prior one, builds an active state where corrections override originals, and injects only the live instruction set into agent context. Prevents model drift when users change their minds mid-session.

23 stars 3 forksAI, LLMs & Agents

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add Parad0x-Labs/openclaw-skills --skill correction-intent-tagger
Manual install
git clone --depth 1 https://github.com/Parad0x-Labs/openclaw-skills.git /tmp/openclaw-skills
mkdir -p ~/.claude/skills
cp -r /tmp/openclaw-skills/plugins/correction-intent ~/.claude/skills/correction-intent-tagger

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Correction vs Instruction Tagger

Detects when a message corrects a prior one and builds an active state
where corrections override originals.

The problem nobody benchmarks: "use Redis" at msg 20, "actually use
Postgres" at msg 47. Standard compression keeps both. This tagger
supersedes Redis, surfaces only Postgres in activeInstructions.

## Usage

```python
from correction_intent import tag_messages, build_active_state, inject_active_state

tagged = tag_messages(messages)
state = build_active_state(tagged)
system_injection = inject_active_state(state)
# inject into system prompt — agent sees CURRENT STATE, not history
```

## What it detects

- CORRECTION: "actually", "scratch that", "instead", "I meant", implicit topic override
- INSTRUCTION: new directive
- ADDITIVE: "also", "additionally" — extends without overriding
- QUERY: question, not actionable
- ACK: "ok", "got it", "understood"

## Correction chain receipt

```python
from correction_intent import correction_chain_receipt
receipt = correction_chain_receipt(state, session_id)
# receipt["chain_root"] is the anchorable 32-byte root; a devnet receipt_anchor is
# available at HSdEQWunzPtNqdzv5HfXuA3zwPLpgTXRyfbndnGamhXs; pass it explicitly to an anchoring client
```

Part of the Parad0x Labs stack: github.com/Parad0x-Labs/openclaw-skills/tree/main/plugins/correction-intent
Read the full skill on GitHub

Details

Skill name
correction-intent-tagger
Path
plugins/correction-intent/SKILL.md
License
MIT
Repo updated
13 November 2026

FAQ

What does the Correction Intent Tagger skill do?

Detects when a user message corrects a prior one, builds an active state where corrections override originals, and injects only the live instruction set into agent context. Prevents model drift when users change their minds mid-session.

How do I install the Correction Intent Tagger skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add Parad0x-Labs/openclaw-skills --skill correction-intent-tagger. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Correction Intent Tagger popular?

It lives in Parad0x-Labs/openclaw-skills, which has 23 GitHub stars and 3 forks.

Related skills

Agent Passport

23

Parad0x-Labs/openclaw-skills

On-chain identity for OpenClaw agents — .null name, ETH↔Solana binding, verifiable agent identity without touching private keys

AI, LLMs & Agents

Context Capsule

23

Parad0x-Labs/openclaw-skills

Compresses older OpenClaw agent session history into a bounded, lane-change-aware context capsule — keeps recent messages verbatim, flags abandoned directions, quarantines injected instructions, and redacts secrets. Local, deterministic, any model.

AI, LLMs & Agents

Dark Null Bridge

23

Parad0x-Labs/openclaw-skills

Archive Dark NULL ZK settlement data into encrypted Liquefy TraceVaults with optional Solana anchoring via an SPL Memo transaction. Bridges zero-knowledge proof outputs into a verifiable, tamper-evident vault.

Other

Liquefy Token Guard

23

Parad0x-Labs/openclaw-skills

Token usage auditor and budget guard for OpenClaw agents — scans traces for waste (duplicate prompts, oversized context, model overkill), writes budgets, and builds a compact context capsule so the next run starts lean.

AI, LLMs & Agents

Openclaw X402 Pay

23

Parad0x-Labs/openclaw-skills

Pay x402-gated APIs and AI agents on Solana — builds an unsigned payment transaction, verifies settlement on-chain, and returns a receipt hash. Hard spend cap enforced before any transaction. BYO signer, no keys held.

AI, LLMs & Agents

AI Mobile Deployment

39

lionelsimai/claude-skills-collection

Deploy AI models to mobile applications. TRIGGERS - Use when user needs help with ai-mobile-deployment related tasks.

AI, LLMs & Agents

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.