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 skills add AIOSAI/AIPass --skill screen_lockgit clone --depth 1 https://github.com/AIOSAI/AIPassWrote 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/aiosai/aipass/screen_lock)<a href="https://agentmods.dev/skills/aiosai/aipass/screen_lock"><img src="https://agentmods.dev/badge/skills/aiosai/aipass/screen_lock/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/aiosai/aipass/screen_lock"><img src="https://agentmods.dev/badge/skills/aiosai/aipass/screen_lock.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00021 | $0.01337 |
| Opus 5.5 | $0.00008 | $0.00535 |
| Sonnet 5.5 | $0.00004 | $0.00267 |
| Haiku 4.5 | $0.00002 | $0.00134 |
Grade A, and why
screen_lock 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 today.
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 — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Screen Lock
What This Does
Password-locks the graphical session and leaves everything running — agents keep working behind the password wall. Nothing sleeps, nothing is killed, no root is needed (no sudoers grant, no polkit rule).
Two paths, tried in order:
loginctl lock-session <id>against the explicitly resolved graphical session. Callers usually run as asystemd --userservice with noXDG_SESSION_ID, so a barelock-sessionhas no ambient session to resolve and may refuse — naming the session is what makes the verb work from there.gdbus→org.gnome.ScreenSaver.Lockas the fallback.
If both fail it says so. A screen that never locked is never reported as locked.
Doctrine — lock is the exception
A destructive action never fires from a locked screen. Lock itself must work from anywhere — that is its whole point.
So this verb is never gated: it does not ask whether the screen is already locked, does not require a desktop environment, and does not refuse when no graphical session can be resolved (it still tries the bare call). Callers that gate other verbs on screen state must leave this one ungated.
The read beside the lock
lock_state() answers whether the screen is locked — the question "did I
lock the laptop?" asked from away (FPLAN-0585). It reads through the same pair
the lock writes through, in the same order, and changes nothing:
loginctl show-session <id> -p LockedHinton the resolved graphical session — the logind mechanismlock-sessionsets, so read and write agree.gdbus→org.gnome.ScreenSaver.GetActivewhen logind cannot answer. (org.freedesktop.ScreenSaver.GetActiveis not implemented on GNOME and is never asked.)
It is not gated either — the lock never is, so its read is not — and it never
raises. Unknown is never unlocked: a read that cannot tell answers
ok: False, locked: None with a reason code, never locked: False.
reason |
Meaning |
|---|---|
no_session |
logind lists no active graphical session of ours, and D-Bus did not answer |
no_reader |
neither loginctl nor gdbus is installed |
read_failed |
a reader ran and failed, or answered something other than yes/no |
What ships with it
5 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.
- today Changed · -2 lines 96807ae36980
- 1mo ago Changed · +45 lines · +7 tokens per session 3e667f4678ab
- 1mo ago First seen · 103 lines · 14 tokens per session scan A 1fea30d8b514
screen_lock is a skill published in the GitHub repository AIOSAI/AIPass (301 stars, last pushed yesterday), licensed MIT. It adds 21 tokens to every session and 1,337 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
heartbeat
Lightweight system health monitoring skill.
clinical-research-session
Run a clinical research study with the researcher from question to results - interview, protocol, approved analysis scripts, and results on the Vitrine board. Use when the user wants to study whether an exposure or treatment affects an outcome, compare patient groups, characterise or model a cohort, write a research…
vitrine-api
Publish results to the researcher's Vitrine board as cards (tables, figures, images and galleries, notes, forms, approvals) and read back what is on it (cards, decisions, form answers, annotations). Use when showing results, asking the researcher to choose or approve, or when they mention a card, a table, the board…
m4-api
Query M4 clinical datasets (MIMIC-IV, MIMIC-IV-Note, MIMIC-IV-ED, eICU, MIMIC-III, MIMIC-CXR-JPG, and demos) from Python with import m4: find tables, run SQL into pandas, search clinical notes, select chest X-ray cohorts and fetch their images, handle M4's errors. Use before writing Python or SQL against M4 data.
mimic-eicu-mapping
Map MIMIC-IV identifiers, tables, time handling and concepts to eICU-CRD as M4 exposes it. Use when translating a MIMIC-IV query to eICU, validating a MIMIC-IV model on eICU, or running one analysis on both databases.
m4-setup
Diagnose and repair an M4 installation. Use when import m4 or import vitrine fails or runs under the wrong Python, a dataset is missing, not ready, or locked, the user's own CSV files are to become a dataset, the BigQuery backend or its credentials are wrong, M4's skills are missing from an agent, or the Vitrine board…