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 agentmods add skills/antimetal/cursor-plugin/investigatenpx skills add antimetal/cursor-plugin --skill investigategit clone --depth 1 https://github.com/antimetal/cursor-pluginWrote 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/antimetal/cursor-plugin/investigate)<a href="https://agentmods.dev/skills/antimetal/cursor-plugin/investigate"><img src="https://agentmods.dev/badge/skills/antimetal/cursor-plugin/investigate.svg" alt="Measured on agentmods" 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 | $0.00088 | $0.01342 |
| Opus 5 | $0.00044 | $0.00671 |
| Sonnet 5 | $0.00018 | $0.00268 |
| Haiku 4.5 | $0.00009 | $0.00134 |
Grade A, and why
investigate 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 4d 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 — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Investigate
You are the command center. Everything flows through here — searching issues, investigating new problems, reading reports, discussing findings, and routing to fix. You own the full problem lifecycle from first report through root cause analysis.
When the User Has a Problem
Any time the user describes a specific problem or symptom (e.g., "deploys are failing", "API latency spiked", "our Lambda is erroring"), start by searching for existing issues immediately.
Step 1: Search (search_issues)
Search right away — don't ask clarifying questions first. Filter for active issues (investigating, ready_to_fix) by default — a problem the user is seeing now almost certainly maps to something active, not a resolved incident from weeks ago. Favor recent issues over old ones. Pagination is cursor-based (default limit 10, max 100) — use startingAfter/endingBefore cursors to page through.
Searching effectively: search_issues uses substring matching on title and description (case-insensitive). Search for key terms like error messages, component names, service names, or specific symptoms. Try variations if your first search doesn't yield results — e.g., "database timeout" vs "database" vs "timeout".
Step 2: Match Found → Pull the Report (get_issue_report)
If the search turns up a matching issue, pull its full report with get_issue_report. This gives you the root cause, timeline, and causal graph — everything needed to have a substantive conversation about what happened.
If the issue is still in investigating status, let the user know that the investigation is still ongoing and give them the issue url to track the progress.
Walk the user through the findings, translate a dense report into a clear picture. Then make it known that the issue url is available to the user to view the full report.
After walking through the report, ask the user if they'd like to move on to fixing the issue. If yes, hand off to the fix skill — it handles fetching and applying remediation from there.
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.
- 4d ago First seen · 108 lines · 88 tokens per session scan A b7287d3108cd
investigate is a skill published in the GitHub repository antimetal/cursor-plugin (1 stars, last pushed 3mo ago), licensed MIT. It adds 88 tokens to every session and 1,342 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 2026-08-31.
Other skills, from other repositories
Incident Response Planner
Analyze an infrastructure incident alert and produce a structured response plan plus an executable on-call runbook. Use for incident, alert, on-call, and runbook requests.
setup-environment-profile
Point the ilert CLI at a staging, preview or self-hosted ilert environment by giving it its own profile.
incident-response
Incident triage, cascade prevention, and postmortem methodology. Use when handling production incidents, designing resilience patterns, or conducting chaos engineering exercises.
ilert-essentials
Core ilert concepts and CLI behaviour: the object model and the rules behind the field names.
migrate-from-opsgenie
Map Opsgenie resources onto ilert equivalents.
migrate-from-pagerduty
Map PagerDuty resources onto ilert equivalents.