Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add 7xuanlu/wenlan --skill run-wenlan-appgit clone --depth 1 https://github.com/7xuanlu/wenlanWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/7xuanlu/wenlan/run-wenlan-app)<a href="https://agentmods.dev/skills/7xuanlu/wenlan/run-wenlan-app"><img src="https://agentmods.dev/badge/skills/7xuanlu/wenlan/run-wenlan-app/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/7xuanlu/wenlan/run-wenlan-app"><img src="https://agentmods.dev/badge/skills/7xuanlu/wenlan/run-wenlan-app.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00071 | $0.05089 |
| Opus 5 | $0.00036 | $0.02544 |
| Sonnet 5 | $0.00014 | $0.01018 |
| Haiku 4.5 | $0.00007 | $0.00509 |
Grade A, and why
run-wenlan-app scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 353 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run wenlan-app (dev)
Tauri 2 desktop app (Rust app/ + React/Vite frontend on :1420). Drive it
with .claude/skills/run-wenlan-app/driver.sh — all paths below are relative
to the unit root (repo or worktree root).
Prerequisites, both Run sections, Gotchas, and Troubleshooting below are
macOS only (driver.sh, vite, ScreenCaptureKit). On Windows the webview is
WebView2 and the recipe is different — jump to
Windows (WebView2).
A debug build refuses to start without a complete isolated runtime identity
(validate_debug_runtime_isolation in app/src/lib.rs), so launch takes the
worktree-scoped config from scripts/dev-runtime.sh and starts its own
daemon on the worktree port. The user's live wenlan-server on :7878 is
never the daemon under test and must never be killed; stop only ends the
daemon this worktree recorded.
Prerequisites
pnpm installdone in the unit root (worktrees start withoutnode_modules).- No sibling checkout needed: the monorepo checkout is the backend, and
scripts/resolve-backend-dir.shprobes the repo root first (an explicitWENLAN_BACKEND_DIRoverrides it; a sibling checkout is only a legacy fallback). The driver defaultsCARGO_TARGET_DIRto the main checkout (a warm cargo cache) viagit rev-parse --git-common-dir, so worktrees work out of the box. - Terminal needs macOS Screen Recording permission (for
shot).
Run (agent path)
.claude/skills/run-wenlan-app/driver.sh build # sidecars + cargo build (decoupled from tauri dev)
.claude/skills/run-wenlan-app/driver.sh launch # isolated daemon + vite up + launch target/debug/wenlan-app directly → "APP UP"
.claude/skills/run-wenlan-app/driver.sh shot /tmp/shot.png # window PNG, works while occluded
.claude/skills/run-wenlan-app/driver.sh stop # kills only the dev app, vite, and this worktree's daemon; never :7878
Logs land in $TMPDIR/wenlan-app.log and $TMPDIR/wenlan-vite.log.
Look at the screenshot after shot — a blank frame means the frontend
did not load.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago Changed · +246 lines · +11 tokens per session 76eabd53f519
- 10d ago First seen · 107 lines · 60 tokens per session scan A d8d03c13e569
run-wenlan-app is a skill published in the GitHub repository 7xuanlu/wenlan (64 stars, last pushed today), licensed Apache-2.0. It adds 71 tokens to every session and 5,089 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
shodh-memory
Persistent memory system for AI agents. Use this skill to remember context across conversations, recall relevant information, and build long-term knowledge. Activate when you need to store decisions, learnings, errors, or context that should persist beyond the current session.
braindb-agent
Persistent memory across sessions via the BrainDB agent. Use at conversation start and whenever you need to recall what you know about the user or save new information to long-term memory.
windows-compat
Audit and harden this Rust repo (code-graph-mcp) for Windows correctness: path-spelling drift between producers, the 32,767-char command-line cap, index-key mismatches, and path predicates that assume one ecosystem's layout. Use whenever touching code that builds, compares, prints, or stores a filesystem path; that…
remnic-memory-workflow
Shared memory workflow for Claude Code agents connected to Remnic — recall before acting, observe during work, remember at the end. Trigger phrases include "what do you remember about", "save this for later", "any context from last time".
deeprefine
Agent-native DeepRefine refinement loop — same control flow as DeepRefine.refine(), graphify search instead of FAISS, session LLM, dry-run review before approved graph writes.
remnic-entities
Browse entities in the Remnic knowledge graph and surface their facts and relationships. Trigger phrases include "tell me about the entity", "look up", "what do we know about".