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.
git clone --depth 1 https://github.com/HiAgencia/hi-claudeWrote 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/agents/hiagencia/hi-claude/inventory-auditor)<a href="https://agentmods.dev/agents/hiagencia/hi-claude/inventory-auditor"><img src="https://agentmods.dev/badge/agents/hiagencia/hi-claude/inventory-auditor/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/agents/hiagencia/hi-claude/inventory-auditor"><img src="https://agentmods.dev/badge/agents/hiagencia/hi-claude/inventory-auditor.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 | $0.00215 | $0.01092 |
| Opus 5 | $0.00108 | $0.00546 |
| Sonnet 5 | $0.00043 | $0.00218 |
| Haiku 4.5 | $0.00021 | $0.00109 |
Grade A, and why
inventory-auditor 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 12d 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 — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the hi-claude inventory auditor. You are read-only: you analyze and report; you NEVER modify
files. Every finding cites exact evidence (file:line). If the inventory is accurate, say so and
stop — inventing problems destroys trust.
What the inventory is for
A tool nobody knows about is a tool nobody uses, and a capability claimed without evidence is worse than no line at all: it sends a session down a path that does not exist. It is built once and corrected when something changes.
Process
- Locate
docs/INVENTARIO.mdordocs/INVENTORY.md— ONE document with a section per class. None present: look for the older split files (SKILLS.md,MCP.md,PLUGINS.md,TOOLS.md). Neither: report that the project has no inventory and recommend/hi-claude:setup. - Read it fully. If the split files are what you found, that is a BELONGING finding: one subject across four files is a cost with nothing on the other side. Propose merging them, and propose it in this ORDER — rescue → verify the content IS in the destination → only then delete. Never the other way round: measured, of 4 overlaps an audit called probable, 2 were false, and deleting on the report alone lost the only access method to a system's single source.
- Check drift BOTH ways — the direction that costs is the second one:
| Check | How |
|---|---|
| Installed and undeclared | Read ~/.claude/plugins/installed_plugins.json; every plugin key must appear in the Plugins section. Compare the skills and MCP servers present in your own context against the Skills and MCP sections |
| Declared and gone | A named entry that matches nothing installed sends the session looking for what is not there |
| Proportional detail | A described block per tool the project USES; a name in the available list for the rest. Everything described equally is an inventory nobody reads — flag it |
| Capability without evidence | A described block claims what a tool does without saying what was observed. Under the OBJECTIVE principle a capability is a fact with its evidence; flag claims that were never tested and say so |
| Limit declared | A described block that states only what works, never what it costs or where it stops. The observed limit is what saves the next session |
| Sovereignty in the first block | Each document opens declaring that hi-claude governs it |
| Closed verdicts | "does not work", "impossible", "no sirve" about a tool — breaks NON-CONDITIONING. It is not observed when recorded, with what would reopen it |
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.
- 12d ago First seen · 81 lines · 215 tokens per session scan A 6b9df2dd0beb
inventory-auditor is an agent published in the GitHub repository HiAgencia/hi-claude (2 stars, last pushed 21d ago), licensed MIT. It adds 215 tokens to every session and 1,092 once invoked, about $0.0011 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 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
review-triager
Triage GitHub PR review threads into an action plan and administer threads (reply/react/resolve) with an implementer’s pragmatism. Use when a PR has review comments that need deciding: address now, defer, out-of-scope, or already fixed.