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/ggemba/squad-mcpWrote 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/commands/ggemba/squad-mcp/inventory)<a href="https://agentmods.dev/commands/ggemba/squad-mcp/inventory"><img src="https://agentmods.dev/badge/commands/ggemba/squad-mcp/inventory/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/commands/ggemba/squad-mcp/inventory"><img src="https://agentmods.dev/badge/commands/ggemba/squad-mcp/inventory.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 | $0.00000 | $0.00918 |
| Opus 5.5 | $0.00000 | $0.00367 |
| Sonnet 5.5 | $0.00000 | $0.00184 |
| Haiku 4.5 | $0.00000 | $0.00092 |
Grade A, and why
inventory 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 — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are running the inventory skill on the user's request:
$ARGUMENTS
Execute the skill exactly as specified at skills/inventory/SKILL.md. The full contract — Inviolable Rules, pack loading, rg sweep, hybrid dispatch, deterministic MD renderer, telemetry — lives there. This file is a thin trigger; the skill file is the source of truth.
Depth (--quick / --normal / --deep)
Same vocabulary as /squad:implement and /squad:review, but mapped to the model strategy for enrichment:
--quick→ Haiku tier-1 only. No escalation. Faster, cheaper, summaries are generic. Best for exploratory first sweep ("how big is the problem?").--normal(default) → Haiku tier-1 + Sonnet escalation for findings withconfidence < 0.7OR rules taggedrequires_semantic: true. Sweet spot.--deep→ Sonnet tier-1 across all findings + Opus for the cross-ref summary section. Use when the inventory will become a canonical migration plan.
If the user passes none, default to --normal.
Output target
Default: ./docs/inventory/<recipe>-<YYYY-MM-DD>.md inside the project. Created if docs/ doesn't exist. Override with --out <path> (accepts Obsidian vault paths, absolute paths, or repo-relative).
If the output path already exists, the skill prompts before overwriting (write-authority gate, same pattern as /squad:grillme).
Dry-run
--dry-run shows the plan (pack rules loaded, file count, bucket count, dispatched-tier breakdown, estimated token cost) without running the LLM enrichment. The rg sweep still runs (it's cheap) and the resulting raw findings are reported in count form only.
Critical reminders
- Reads the codebase; writes ONE file. The skill produces a single MD report at the user-supplied (or default) path. It never edits source code, never modifies git state, never commits, never pushes.
- Pack loading is strict. A malformed YAML pack fails the run at Phase 1 — do not partially run with a broken pack.
- Cross-tier escalation is binding. Findings from rules with
requires_semantic: trueMUST go to Sonnet+ (or skip enrichment entirely if--quick). Never silently downgrade. - Determinism is the contract. Same JSON intermediate → byte-identical MD (
src/inventory/render-md.tsenforces this via Zod parse and stable sort). If you find yourself adding wall-clock-dependent content, stop — that breaks the contract. - No AI attribution in the generated MD. The footer says "Generated by squad-mcp inventory", not "Generated with AI".
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 · 41 lines · 0 tokens per session scan A 884a6b5b49a3
inventory is a command published in the GitHub repository ggemba/squad-mcp (4 stars, last pushed 2mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 918 tokens. 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 commands, from other repositories
db
Diagnose and fix a database bottleneck evidence-first — EXPLAIN, indexes, N+1, locks, pooling, safe migrations.
debug
Systematic debugging: root cause before any fix, one hypothesis at a time, and a three-strikes rule that questions the architecture instead of stacking patches.
inconsistences
TODOs, errors, incosistencies, potential code and typesafety violations in.
sanity-review
Review code for Sanity best practices and common issues.
recover
Diagnose and recover from interrupted or corrupted state.
analyze-misfires
Identify skills injected where not needed, propose regex and description tightening.