Borrowing it
Nothing to install: this file belongs to capad-xyz/searchts. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/capad-xyz/searchts/main/.agents/skills/verify-searchts/SKILL.mdgit clone --depth 1 https://github.com/capad-xyz/searchtsWrote 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/capad-xyz/searchts/verify-searchts)<a href="https://agentmods.dev/skills/capad-xyz/searchts/verify-searchts"><img src="https://agentmods.dev/badge/skills/capad-xyz/searchts/verify-searchts/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/capad-xyz/searchts/verify-searchts"><img src="https://agentmods.dev/badge/skills/capad-xyz/searchts/verify-searchts.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 | $0.00076 | $0.01301 |
| Opus 5.5 | $0.00030 | $0.00520 |
| Sonnet 5.5 | $0.00015 | $0.00260 |
| Haiku 4.5 | $0.00008 | $0.00130 |
Grade C, and why
verify-searchts scanned grade C with 1 finding 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 17d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
Teardown is `rm -rf .venv` and nothing else, and only if this run created it. How it starts
The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verify searchts
searchts is a CLI a user runs in a terminal. Every drive is a fresh process that exits on its own: there is no server to start, no browser to attach to and no PTY to hold open, because none of the user-facing verbs is interactive. The proof is the process's own output, its exit code, and the file it wrote.
Read features/README.md first. It holds the baseline
preconditions, the driving conventions, the proof standard and the contract every
feature file follows.
Launch
There is no long-lived process to launch. Launch means make the interpreter exist once, then each drive is a single command.
uv venv .venv && uv pip install -c constraints.txt -e ".[dev]"
.venv/bin/python -m searchts --version
Ready when that prints searchts vX.Y.Z. Use .venv/bin/python -m searchts, not
the searchts script on PATH: the PATH copy may be an older global install,
and a proof against the wrong build is worse than no proof. The documented
install is the one in CLAUDE.md; uv and pip both work.
Teardown is rm -rf .venv and nothing else, and only if this run created it.
Doctor
Run this before driving anything, and again after any drive that misbehaved.
.venv/bin/python -m searchts doctor --json
Exit 0 means the command ran. It does not mean every channel is green: optional
platform CLIs report "status": "warn" and that is the expected state on a
machine without them. Worth driving is: github, youtube, tiktok,
instagram, reddit-video and rss all "status": "ok" with a non-null
active_backend.
web is the one that changes what a run can prove. Read its message: it names
every tier of the read ladder that is missing, and a missing tier is why a
bot-walled page cannot be fetched on this machine.
Drive
Three rules keep the evidence honest.
- Capture to a file, never pipe.
searchts grab … | headmakes Python writeBrokenPipeError: [Errno 32] Broken pipeto stderr, and the artifact becomes the pipe rather than the tool. Redirect stdout and stderr separately. - Use
--jsonwherever the verb has it.read,search,grabanddoctordo;transcribeandgetdo not and write files instead. Human output isrich-formatted and moves with terminal width, so it is not a stable thing to assert on. - Keep the streams apart. stdout is the answer, stderr is progress. Ticks do not go to stdout, so a drive that asserts on stdout alone is asserting on the payload.
What ships with it
8 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.
- 17d ago First seen · 116 lines · 76 tokens per session scan C 07e3d7a4e7b7
verify-searchts is a skill published in the GitHub repository capad-xyz/searchts (2 stars, last pushed yesterday), licensed MIT. It adds 76 tokens to every session and 1,301 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-13.
Other skills, from other repositories
test-writer
Expert test writer for unit, component, E2E, and integration tests. Writes only high-value tests — branching, side effects, transactions, corner cases. Use for TDD or filling coverage gaps. Never writes tests for constants, single pass-throughs, or thin wrappers. Adapts to the project's test framework and conventions.
e2e-testing
Guide for running end-to-end tests of the Qwen Code CLI, including headless mode, MCP server testing, and API traffic inspection. Use this skill whenever you need to verify CLI behavior with real model calls, reproduce user-reported bugs end-to-end, test MCP tool integrations, or inspect raw API request/response…
terminal-capture
Automates terminal UI screenshot testing for CLI commands. Applies when reviewing PRs that affect CLI output, testing slash commands (/about, /context, /auth, /export), generating visual documentation, or when 'terminal screenshot', 'CLI test', 'visual test', or 'terminal-capture' is mentioned.
tmux-real-user-testing
A real-user test workflow for Qwen Code's terminal interface, using tmux to drive the program and capture readable screen snapshots. tmux is a tool for running and observing terminal sessions.
agent-reproduce-align
Use after a Codex or Claude Code feature has been implemented in Qwen Code to run the selected reference agent and Qwen Code under the same scenario, capture HTTP and terminal traces, compare request bodies, tool/function schemas, outputs, and iterate until the reproduced behavior is close enough.
verify
Exercise the real app/API/CLI and collect observable evidence; tests alone do not count as end-to-end verification.