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/rokoss21/iosm-cli/valid-skillnpx skills add rokoss21/iosm-cli --skill valid-skillgit clone --depth 1 https://github.com/rokoss21/iosm-cliWhat 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 | $0.00011 | $0.00033 |
| Opus 5 | $0.00005 | $0.00016 |
| Sonnet 5 | $0.00002 | $0.00007 |
| Haiku 4.5 | $0.00001 | $0.00003 |
Grade A, and why
valid-skill 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 2d 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.
This is a copy
100% identical to valid-skill — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
What it actually says
Valid Skill
This is a valid skill that follows the Agent Skills standard.
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.
- 2d ago First seen · 9 lines · 11 tokens per session scan A db6af8bf9364
valid-skill is a skill published in the GitHub repository rokoss21/iosm-cli (146 stars, last pushed 4mo ago), licensed MIT. It adds 11 tokens to every session and 33 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to valid-skill, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
manta-pr-workflow
The standard implementer workflow for MANTA (Better UI) agents — checkout, branch, implement, typecheck + test, push, draft PR, verification results template, reassign to reviewer. Plus the 50-turn self-policed budget cap. Load this when you're about to take action on an assigned implementer issue.
manta-e2e-smoke
Drive MANTA's built Electron app in a real renderer context (Playwright's electron launcher) and assert that key UI surfaces render correctly — no crash, no blank screen, sidebar/chat/terminal present. Load BEFORE marking any frontend/UI task done, and when manta-pr-workflow or manta-handle-reviewer-return verifies a…
manta-handle-reviewer-return
Protocol for MANTA implementer agents when reviewer reassigns an issue back. Distinguishes Block (fix-here) vs Block (followup-issue) and handles Question resolutions. Load this when an issue you previously sent to reviewer comes back to you with status todo.
verify-build-manta
Runs npm run typecheck && npm test. Mandatory before marking any task done.
manta-token-economy
Token-economy discipline for MANTA agents on OpenCode runtime — batch parallel tool calls, compound shell commands, avoid re-reading after Edit, use offset/limit on large files. Load this if a task touches a long file or you find yourself making many sequential tool calls.
manta-live-session
Talk to a real manta box from an agent session — create a chat session, send a prompt, read the reply, and mint a pairing code for a phone or Simulator. Use whenever a task needs live box data rather than a unit test. Always runs on voska/default.