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 archetypeai/agent-skills --skill atai-task-verification-agentgit clone --depth 1 https://github.com/archetypeai/agent-skillsWrote 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/archetypeai/agent-skills/atai-task-verification-agent)<a href="https://agentmods.dev/skills/archetypeai/agent-skills/atai-task-verification-agent"><img src="https://agentmods.dev/badge/skills/archetypeai/agent-skills/atai-task-verification-agent/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/archetypeai/agent-skills/atai-task-verification-agent"><img src="https://agentmods.dev/badge/skills/archetypeai/agent-skills/atai-task-verification-agent.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 | $0.00216 | $0.08555 |
| Opus 5 | $0.00108 | $0.04278 |
| Sonnet 5 | $0.00043 | $0.01711 |
| Haiku 4.5 | $0.00022 | $0.00856 |
Grade A, and why
atai-task-verification-agent 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 8d 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 — 529 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TVA Agent — Managed Task Verification via the Agents API
The TVA agent checks a recording against a procedure you supply. You hand the platform an .mp4 and an SOP as a .txt; it samples frames, transcribes the audio, fuses both with your procedure in one pass, and returns one verdict per SOP step:
video ──► sample frames ──► preprocess ─┐
├─► newton-fusion f1-0 ─► parse ─► per-step verdicts
SOP (.txt) ──► PrepareSOPNode ──────────┤
whisper ASR ───────────────────┘
TVA V1 is zero-shot. The tva blueprint pins its own models (newton-fusion:1.0 and whisper:large-v3), so like mga — and unlike osm/red — there is no classifier to fit and no artifacts map to pass.
When to Apply
Use when the procedure is already known and the question is whether it was followed: did the operator install the o-ring, torque the fitting, apply the tape? Each verdict carries a status, a time range and a reason.
Do not treat that output as an audit trail. A PASSED verdict is not evidence the step happened, and the reason is a restatement of your SOP rather than a report of the video — measured, with the frames to prove it, in §1 below. Use this to triage which recordings a human should watch, not to sign work off unreviewed.
Do NOT use when:
| Need | Use instead |
|---|---|
| No procedure exists yet — extract one from the video | atai-manual-generation-agent (mga) |
| One question about a clip, stateless | atai-newton-fusion-model (/query) |
| Classify sensor windows into operational states | atai-operational-state-monitoring-agent |
| Detect a rare event in a long recording | atai-rare-event-detection-agent |
TVA vs MGA in one line: MGA writes the procedure, TVA checks against one. That difference is visible in the blueprints — and it is the reason TVA accepts a runtime input MGA does not:
# mga # tva
source: source:
key: MultiModalSource key: MultiModalSource
config: config: {} # accepts a .txt
default_text: ${values.prompt}
text_extensions: [] # NO text input
What ships with it
13 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.
- references/.env.example 542 B
- references/requirements.txt 91 B
- references/run_tva_agent.py 27 KB runs code
- references/sample_data/1_fail_2_pass_3_pass_A.mp4 7710 KB
- references/sample_data/1_pass_2_pass_3_fail_A.mp4 5309 KB
- references/sample_data/1_pass_2_pass_3_pass_A.mp4 7643 KB
- references/sample_data/oring-numbered.txt 276 B
- references/sample_data/README.md 6.0 KB
- references/sample_data/tva-output-1_fail_2_pass_3_pass_A-FALSE-PASS.json 691 B
- references/sample_data/tva-output-1_pass_2_pass_3_fail_A-CORRECT-MISSING.json 834 B
- references/sample_data/tva-output-1_pass_2_pass_3_fail_A-mnt2048-EMPTY.json 44 B
- references/sample_data/tva-output-1_pass_2_pass_3_pass_A.json 678 B
- tests/test_references.py 16 KB runs code
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.
- 8d ago First seen · 529 lines · 216 tokens per session scan A 982de8728b14
atai-task-verification-agent is a skill published in the GitHub repository archetypeai/agent-skills (5 stars, last pushed 18d ago), licensed Apache-2.0. It adds 216 tokens to every session and 8,555 once invoked, about $0.0011 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
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
skill-benchmark
Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.
playwright-cli
Automates browser interactions for testing and validating your own web applications using playwright-cli. Use when you need terminal-first browser control for navigation, form filling, screenshots, tracing, bound browser sessions, debugging, or generating Playwright test code. Only use against applications you own or…