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 starbringer/AI-Insights --skill ai-usage-reviewgit clone --depth 1 https://github.com/starbringer/AI-InsightsWrote 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/starbringer/ai-insights/ai-usage-review)<a href="https://agentmods.dev/skills/starbringer/ai-insights/ai-usage-review"><img src="https://agentmods.dev/badge/skills/starbringer/ai-insights/ai-usage-review/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/starbringer/ai-insights/ai-usage-review"><img src="https://agentmods.dev/badge/skills/starbringer/ai-insights/ai-usage-review.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 | $0.00156 | $0.01339 |
| Opus 5 | $0.00078 | $0.00669 |
| Sonnet 5 | $0.00031 | $0.00268 |
| Haiku 4.5 | $0.00016 | $0.00134 |
Grade A, and why
ai-usage-review 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 — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI usage review
Turn recorded usage into a small number of changes that are worth making.
Two rules govern the whole review:
- Evidence before advice. Every finding cites a number this session actually measured. No generic best-practice lists.
- You never edit config as part of the review. The MCP server is read-only by design. Report findings, then apply only what the user picks, using your normal edit tools so each change goes through their approval.
1. Establish scope
Ask nothing you can look up. Run list_providers first, then pick the provider:
- The user named one → use it.
- Exactly one has
hasData: true→ use it, say which. - Several do → ask which, or use
provider: "all"if they want the whole picture.
Pass that value as provider on every subsequent call. Default is Claude Code.
If the tools are missing, the dashboard is not running: tell the user to run
bun run start in the AI Insights directory, and stop.
2. Gather
Read references/playbook.md now — it holds the thresholds, the diagnosis rules
and the exact fix text for each finding. Then collect, in this order:
| Step | Tools | What you are looking for |
|---|---|---|
| Shape of spend | get_usage_summary, get_model_usage, get_project_usage |
Scale, trend, model mix, where it lands |
| Waste signals | get_usage_timeseries, get_top_runs, get_top_turns |
Spikes, cache collapse, outlier calls |
| Per-session detail | get_run_usage on the top 2–3 runs |
Bucket split (base/MCP/skills/sub-agents) and its built-in advice |
| Always-on context | list_instruction_files, then read_instruction_file on the largest |
Per-turn tax paid on every single call |
| Extension ROI | list_skills, get_skill_usage, list_commands, list_mcp_servers, get_mcp_usage |
Cost carried vs. value returned |
| Determinism gaps | list_hooks, get_dependency_graph |
Rules that should be hooks; hooks that never fire |
| Correctness | get_effective_config, get_permissions, list_memory_stores |
Ignored layers, thin allowlists, orphaned memory |
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.
- 12d ago First seen · 104 lines · 156 tokens per session scan A af1623006377
ai-usage-review is a skill published in the GitHub repository starbringer/AI-Insights (2 stars, last pushed 23d ago), licensed MIT. It adds 156 tokens to every session and 1,339 once invoked, about $0.0008 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 skills, from other repositories
ci-self-heal
A diagnostic guide for analysing failed continuous-integration checks. Continuous integration, or CI, automatically tests and builds code changes; this guide classifies failures and suggests fixes without editing the code.
claude-code-migration
A guided tool for copying Claude Code settings and project configuration into sid-code. It checks what can be moved, asks before writing, and records what was transferred.
code-review
A review process for a pull request, which is a proposed set of code changes. It examines the changed code and its surrounding files for bugs, security problems, design issues, and lint errors.
security-audit
A security review skill for examining code changes in a pull request, which is a proposed set of changes awaiting approval. It reports issues such as injection, exposed secrets, unsafe authentication, weak cryptography, vulnerable dependencies, and infrastructure mistakes.
incident-rca
A diagnostic process for producing a structured root-cause analysis of a serious production incident, such as a P0 or P1 outage. It compares logs, measurements, request traces, code changes, and architecture decisions, and gives ranked explanations and actions.
code-governance
A review skill that checks pull-request changes for software licenses, personal information, compliance details, and an audit trail. An audit trail is a record showing what was checked and why.