# Lexi

> >

- **Type:** Skill
- **Install:** `agentstack add skill-jazz1x-honne-lexi`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [jazz1x](https://agentstack.voostack.com/s/jazz1x)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [jazz1x](https://github.com/jazz1x)
- **Source:** https://github.com/jazz1x/honne/tree/main/skills/lexi
- **Website:** https://skills.sh/jazz1x/honne

## Install

```sh
agentstack add skill-jazz1x-honne-lexi
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# lexi — Lexicon Axis

Standalone lexicon extraction and analysis. Uses the unified `honne` CLI, presents findings, records claims.

## Step 1: HITL (scope + locale)

Ask for scope (repo/global) AND locale (ko/en/jp). Set `SCOPE` and `LOCALE` variables from the replies.

## Step 2: Scan

`bash "${CLAUDE_PLUGIN_ROOT}/scripts/honne" scan --base-dir ".honne" --scope "$SCOPE"`

## Step 3: Axis extract

`bash "${CLAUDE_PLUGIN_ROOT}/scripts/honne" axis run lexicon --scan .honne/cache/scan.json --locale "$LOCALE"`

## Step 4: Review JSON

Read JSON output — review `candidate_claim`, `quotes`, `insufficient_evidence` fields.

## Step 5: HITL accept/reject/edit

Present candidate claim with sample quotes, ask (y/n/edit):
- **y**: proceed to Step 6
- **n**: record rejection — `bash "${CLAUDE_PLUGIN_ROOT}/scripts/honne" record claim --base-dir ".honne" --type rejection --axis lexicon --scope "$SCOPE" --text "$candidate"`; skip to done
- **edit**: user provides revised text, use as claim

## Step 6: Record claim

`bash "${CLAUDE_PLUGIN_ROOT}/scripts/honne" record claim --base-dir ".honne" --type claim --axis lexicon --scope "$SCOPE" --text "$claim"`

## Step 7: Check past rejections

Avoid re-proposing: `bash "${CLAUDE_PLUGIN_ROOT}/scripts/honne" query --base-dir ".honne" --type rejection --tag lexicon --scope "$SCOPE"`

**Progress monitoring** — use Monitor until-loop (never `sleep N && cat`):
```bash
# ✓ Monitor: until [ -f ".honne/cache/.axis_lexicon.json" ]
```

Report saved to stdout + .honne/assets/claims.jsonl

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [jazz1x](https://github.com/jazz1x)
- **Source:** [jazz1x/honne](https://github.com/jazz1x/honne)
- **License:** MIT
- **Homepage:** https://skills.sh/jazz1x/honne

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-jazz1x-honne-lexi
- Seller: https://agentstack.voostack.com/s/jazz1x
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.
