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/ajmalaksar25/searchlight/skillnpx skills add ajmalaksar25/searchlight --skill skillgit clone --depth 1 https://github.com/ajmalaksar25/searchlightWrote 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/ajmalaksar25/searchlight/skill)<a href="https://agentmods.dev/skills/ajmalaksar25/searchlight/skill"><img src="https://agentmods.dev/badge/skills/ajmalaksar25/searchlight/skill.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 | $0.00124 | $0.02484 |
| Opus 5 | $0.00062 | $0.01242 |
| Sonnet 5 | $0.00025 | $0.00497 |
| Haiku 4.5 | $0.00012 | $0.00248 |
Grade A, and why
searchlight 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 3d 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Searchlight
Run a website from "I don't know what's wrong" to "fixed and verified", using the searchlight
MCP server for data + provisioning and the agent's repo access for fixes. Meet the user where
they are (beginner → pro). Never act without confirming first.
Commands
Route on the first word of the argument; everything after is context (usually a site).
| Invocation | Does | Stops at |
|---|---|---|
/searchlight audit [site] |
Read-only diagnosis: detect + explain, triaged. No changes, no provisioning, no deploy. | end of §2, then offer to fix |
/searchlight setup [site] (or no arg) |
The full guided loop: interview → detect → confirm → provision → fix → deploy → verify | §5 |
/searchlight fix [site] |
Assume already audited; go straight to plan → confirm → fix → deploy → verify | §5 |
audit is the safe default to lead with for a nervous first-time user: show them the picture,
then ask whether to fix. setup / fix change real things and always confirm first.
0. Prerequisites (check, don't assume)
- The
searchlightMCP server must be connected. Callauth_status/setup_status.- Not authenticated → tell them to run
searchlight login --setup, or use theauth_logintool. - For the full hands-off run, the server must be in setup mode (
setupMode: true, started with--setup) and the token must include the setup scopes (re-login with--setup).--setupis self-sufficient: it turns on provisioning (GA4/GTM/verification) and Search Console writes (sitemap submit) and reportswriteEnabled: true. IfsetupModeis false or scopes are missing, say so up front and give the exact one-time command before proposing any write/provision step — don't discover it mid-run. (Foraudit, read-only auth is enough; you don't need setup mode.)
- Not authenticated → tell them to run
- If the user's site code is open in this workspace, you can apply fixes. If not (hosted CMS), you give exact instructions instead of editing.
1. Interview FIRST (ask, confirm — don't barrel ahead)
(For audit, keep this to one line: which site? Then go straight to §2 and stop after.)
Ask only what you need, in plain language, and confirm before doing anything:
- Which site? (
list_sites; resolve an alias or pick.) - What's the goal? (rank a new site / fix indexing / add analytics / general audit.)
- Analytics: do they want it? Create a new GA4 property or connect an existing one? (
ga_list_propertiesto see what exists.) Under which account? (list_ga_accounts.) Searchlight installs the GA4 tag directly via the gtag snippet. - What do they want to track? (key events/conversions — keep it simple for beginners.)
- Code in this repo, or hosted?
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.
- 3d ago First seen · 93 lines · 124 tokens per session scan A 50b67e90a1e1
searchlight is a skill published in the GitHub repository ajmalaksar25/searchlight (0 stars, last pushed 2mo ago), licensed MIT. It adds 124 tokens to every session and 2,484 once invoked, about $0.0006 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…