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 Caxson/proof-of-work --skill verified-deliverygit clone --depth 1 https://github.com/Caxson/proof-of-workWrote 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/caxson/proof-of-work/verified-delivery)<a href="https://agentmods.dev/skills/caxson/proof-of-work/verified-delivery"><img src="https://agentmods.dev/badge/skills/caxson/proof-of-work/verified-delivery/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/caxson/proof-of-work/verified-delivery"><img src="https://agentmods.dev/badge/skills/caxson/proof-of-work/verified-delivery.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 | $0.00074 | $0.00581 |
| Opus 5 | $0.00037 | $0.00291 |
| Sonnet 5 | $0.00015 | $0.00116 |
| Haiku 4.5 | $0.00007 | $0.00058 |
Grade A, and why
verified-delivery 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 12d 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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verified Delivery
Turn the request into a verification-backed chain. Do not claim success without proof.
Procedure
-
Locate & read. Scope the relevant code and logs — use search to bound the area; do NOT read an entire large repo. Understand the real interaction logic. If a design or assumption looks wrong, STOP and confirm with the user before building.
-
Decompose, and attach a verification to each step. For every step, decide up front HOW it will be proven, by surface:
- Web / UI → drive it with Playwright and screenshot the result.
- Data / backend → write a script that calls the REAL service, then query the DB/SQL to confirm the write actually landed.
- Blocked by hardware/device → close the loop via API first, and tell the user this is a stand-in for the real device test.
- Mobile → drive the device and screenshot.
- Always use REAL data. Never mock data unless the user explicitly asks for mock data.
-
Confirm the plan (for non-trivial work). Send the user the step plan plus how each step will be verified, before writing code. Obvious or small tasks may proceed directly with a one-line note.
-
Build and verify step by step. Each step must be proven by a REAL action on REAL data — not a simulated action and not an asserted success.
-
On repeated failure (three or more). Stop. Give a full cause analysis that FIRST rules out your own bug. Do NOT fake a pass and do NOT degrade or work around the problem. Re-plan from a new angle.
-
End-to-end proof. When everything is done, run one end-to-end verification. Save the evidence chain + screenshots to a TEMP directory (never the user's home root). In chat, give only a compact table: step → how verified → result → evidence path. The task is "done" ONLY after the user confirms.
Leverage
- Dispatch an INDEPENDENT subagent for adversarial verification — let it try to DISPROVE that a step works, rather than confirm it.
- For multi-step or parallel verification, orchestrate it as a workflow.
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.
- 12d ago First seen · 51 lines · 74 tokens per session scan A 562575bc338f
verified-delivery is a skill published in the GitHub repository Caxson/proof-of-work (2 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 74 tokens to every session and 581 once invoked, about $0.0004 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".
neuron-evaluation-engineer
Create and run AI evaluations with datasets, assertions, and output drivers in Neuron AI. Use this skill whenever the user mentions evaluation, testing AI systems, creating evaluators, dataset-driven testing, assertion-based validation, or wants to measure AI system performance. Also trigger for tasks involving…
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.