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 commands/dork-labs/dorkos/self-testgit clone --depth 1 https://github.com/dork-labs/dorkosWrote 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/dork-labs/dorkos/self-test)<a href="https://agentmods.dev/commands/dork-labs/dorkos/self-test"><img src="https://agentmods.dev/badge/commands/dork-labs/dorkos/self-test.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 | $0.00038 | $0.07879 |
| Opus 5 | $0.00019 | $0.03939 |
| Sonnet 5 | $0.00008 | $0.01576 |
| Haiku 4.5 | $0.00004 | $0.00788 |
Grade F, and why
self-test scanned grade F with 4 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -s "http://localhost:$API_PORT/api/config" | python3 -c " Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
JSONL_FILE=$(find ~/.claude/projects -name "${URL_SESSION_ID}.jsonl" -type f 2>/dev/null) Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
dotenv -- sh -c 'rm -rf /tmp/dorkos-test-mode-4243 && turbo run build --filter=@dorkos/server && pnpm --filter @dorkos/server exec tsx src/index.ts' Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -sf -X PATCH http://localhost:4248/api/config -H 'content-type: application/json' \ How it starts
The opening of the file, as written. The whole thing — 613 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Self-test the DorkOS chat UI in a live browser session. This command drives real interactions through the full stack, monitors JSONL transcripts on disk, compares API vs UI state at every step, researches any issues found, and produces an evidence-based findings report. If bugs or significant UX issues are found, it generates a prompt for the /flow:ideate command (requires the flow plugin, dork-labs/marketplace, loaded via --plugin-dir).
Argument Parsing
Parse $ARGUMENTS for two optional inputs:
-
URL: Any argument starting with
http— use asTEST_URL. Default (expand$HOMEto the actual home directory; create the dir if missing withmkdir -p "$HOME/Keep/temp/empty"):TEST_URL="http://localhost:6241/?dir=$HOME/Keep/temp/empty" -
Focus areas: An argument starting with
focus:— comma-separated list of specific areas to test. Examples:focus:streaming— Focus on SSE streaming, freeze detection, chunk deliveryfocus:history— Focus on reload-from-history, message persistence, JSONL fidelityfocus:tasks— Focus on task list UI (TaskCreate), task state renderingfocus:tools— Focus on tool call cards, approval flows, expand/collapsefocus:scroll— Focus on auto-scroll, viewport overflow, scroll anchoringfocus:sidebar— Focus on session list, new session, session switchingfocus:status— Focus on status bar, model selector, permission modefocus:code— Focus on code block rendering, syntax highlighting, copy buttonfocus:commands— Focus on slash command palette, command discoveryfocus:markdown— Focus on markdown rendering, links, lists, headings
Multiple areas can be combined:
focus:streaming,history,tasksWhen focus areas are specified:
- Phase 4 messages are tailored to exercise those areas specifically
- Phase 5b checks are weighted toward those areas
- Phase 6 research digs deeper into focused areas
- Unfocused areas still get basic coverage but with less depth
When no focus is specified, run the full default test suite.
-
mode:sandbox|live— which stack the run drives (meta/chat-capabilities.md§11).mode:sandbox— the test-mode runtime: throwaway data dir, deterministic, no model spend. Verifies UI plumbing (rendering, history reload, stream lifecycle).mode:live— the dev stack with a real runtime. Model defaults toclaude-haiku-4-5(Phase 3). Verifies streaming feel, real tool loops, timing.- If the invocation does not state a mode, ASK the user before spending anything (
AskUserQuestion, offering both, with the cost of each). Never assumelive.
Store parsed values as TEST_URL, FOCUS_AREAS (array, possibly empty), and MODE.
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 · 613 lines · 38 tokens per session scan F 07d8b81f4ade
self-test is a command published in the GitHub repository dork-labs/dorkos (9 stars, last pushed 3d ago), licensed MIT. It adds 38 tokens to every session and 7,879 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it F with 4 findings (downloads and executes remote code, reads agent configuration directories, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
dark-factory-behavioral-e2e
Holdout-pattern E2E validator. Drives agent-browser against the running DynaChat app to verify that the PR's user-facing behavior actually matches what the linked issue asked for.
testarch-framework
Initialize production-ready test framework architecture (Playwright or Cypress) with fixtures, helpers, and configuration.
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
test-run
Run all tests and report coverage to @chief.
playwright-cli
Use this skill whenever the user wants to automate a browser, scrape web content, take screenshots or PDFs of pages, fill out forms, click through UI flows, or run end-to-end tests — without using an MCP server. This skill drives Playwright directly from the bashtool via Node.js scripts. Trigger whenever the user says…
qa
Dispatch a verifiable browser task to the qa-tester subagent. Use for regression checks, smoke tests, and any task with a clean pass/fail outcome. Supports an EXHAUSTIVE mode that verifies every control behind an honesty gate.