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 viknesh20-20/claude-code-tool-kit --skill reference-appgit clone --depth 1 https://github.com/viknesh20-20/claude-code-tool-kitWrote 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/viknesh20-20/claude-code-tool-kit/reference-app)<a href="https://agentmods.dev/skills/viknesh20-20/claude-code-tool-kit/reference-app"><img src="https://agentmods.dev/badge/skills/viknesh20-20/claude-code-tool-kit/reference-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/viknesh20-20/claude-code-tool-kit/reference-app"><img src="https://agentmods.dev/badge/skills/viknesh20-20/claude-code-tool-kit/reference-app.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00054 | $0.01435 |
| Opus 5.5 | $0.00022 | $0.00574 |
| Sonnet 5.5 | $0.00011 | $0.00287 |
| Haiku 4.5 | $0.00005 | $0.00144 |
Grade A, and why
reference-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 1mo 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 — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/reference-app
Point Claude at another fully-developed application the user wants to use as a reference. Build a structured understanding of it. Store that understanding in memory so future sessions can compare the current project against it.
When to use
- The user is starting a new project and wants it modeled on one they've already built.
- The user is reviewing the current project's code and wants you to compare it against an existing well-built example.
- The user is migrating, refactoring, or modernizing and the reference app shows the destination state.
- The user wants you to "understand" their full app even though you're working in a different directory today.
Operating method — ASK FIRST, THEN ANALYZE
Step 0: Confirm intent.
Before reading anything, ask:
- "What would you like me to learn from this reference app?" — pattern conventions / architecture / specific module / all of it.
- "What does the current project share with the reference?" — same domain / same stack / similar feature / different / unsure.
- "Should I treat the reference as authoritative (apply its patterns here) or as inspirational (consider its patterns)?"
These three answers shape the depth and tone of the resulting memory. Don't skip them.
Step 1: Validate the path.
- Confirm the path exists and is readable.
- Check it has a recognizable project structure (package.json, pyproject.toml, go.mod, Cargo.toml, etc.). If not, ask the user what kind of project it is.
Step 2: High-level scan (light pass).
Read in this order, one file each, no deep dives yet:
- Top-level README.
CLAUDE.mdif it exists.- The package/dependency manifest.
- The deployment / CI config.
- The directory tree (one level deep, maybe two).
Step 3: Architecture map.
- Identify the entry points (server start, client root, CLI commands).
- Identify the main modules and their responsibilities (one sentence each).
- Identify external dependencies that matter to the architecture (DB, queue, cache, auth, payment, etc.).
- Identify the testing approach (framework, layering).
- Identify the deploy target.
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.
- 1mo ago First seen · 171 lines · 54 tokens per session scan A fa5c57a8a383
reference-app is a skill published in the GitHub repository viknesh20-20/claude-code-tool-kit (7 stars, last pushed 5mo ago), licensed MIT. It adds 54 tokens to every session and 1,435 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-08-31.
Other skills, from other repositories
wiki-ingest
Use when an operation issue asks to ingest a captured raw/ source into the LLM Wiki, or the user says "ingest ". Create durable source, entity, concept, synthesis, index, and log pages; use paperclip-distill for Paperclip bundles.
wiki-lint
Use when an LLM Wiki operation issue is a lint or health check. Audit for contradictions, orphans, weak provenance, broken links, missing concept pages, and index/log drift; return triage findings without auto-fixing.
para-memory-files
Use a file-based PARA memory system to store, retrieve, and organize durable knowledge across sessions. Trigger on saving facts, daily notes, entity records, weekly synthesis, recall, tacit user patterns, or plan memory.
handoff
Write a compact, decision-ready handoff so the next session (or the user) can continue without reconstructing the current one. Use when the session is ending, context is running low, the user asks for a handoff / 'pass the baton' / 'hand off', or a long-running operation needs a durable state checkpoint.
forget
Help the user remove a stored fact using Codewhale's Context Lens when they ask to forget or delete a memory.
claude-compaction-restore
Use when a Claude Code session is about to compact, has just compacted, reached its context limit, resumed after /compact, or must rebuild its working picture of a long-running seat from its own restore map, its JSONL transcript and the files it touched.