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 sanctifai/skills --skill proof-of-humangit clone --depth 1 https://github.com/sanctifai/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/sanctifai/skills/proof-of-human)<a href="https://agentmods.dev/skills/sanctifai/skills/proof-of-human"><img src="https://agentmods.dev/badge/skills/sanctifai/skills/proof-of-human/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/sanctifai/skills/proof-of-human"><img src="https://agentmods.dev/badge/skills/sanctifai/skills/proof-of-human.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 | $0.00083 | $0.01733 |
| Opus 5 | $0.00042 | $0.00866 |
| Sonnet 5 | $0.00017 | $0.00347 |
| Haiku 4.5 | $0.00008 | $0.00173 |
Grade A, and why
proof-of-human 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.
How it starts
The opening of the file, as written. The whole thing — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Proof of Human
SanctifAI Trust Chat bridge — one of three Trust surfaces (Embedded, Chrome
extension, Chat bridge). All three produce the same participation +
certificate_url. This skill is the chat path: you cannot run WebAuthn
in-card, so you mint here, a human opens Chrome, and you poll for the cert.
Default APP_BASE_URL: https://bridge.trust.sanctifai.com
Always use the approve_url the API returns. Never invent localhost links.
The agent never holds TRUST_API_KEY — the hosted bridge does. Self-host this
plugin only if you need a custom allowlisted origin.
Embedded and Extension: https://trust.sanctifai.com/agents/trust/skill.md
Flow
1. POST {APP_BASE_URL}/api/v1/attestations
2. Give the human approve_url — they open it in Chrome
3. Poll GET {APP_BASE_URL}/api/v1/attestations/{id}
or GET {APP_BASE_URL}/api/v1/attestations/{id}/wait
4. When status is completed, paste certificate_url
1. Create
POST {APP_BASE_URL}/api/v1/attestations
{
"task_type": "GEN",
"domain": "GEN",
"task_subtype": "Chat approval",
"requesting_agent": "Sebastian - Dev Bot",
"taskData": { "summary": "opaque task payload" },
"resultData": { "decision": "approved" }
}
task_type and domain are 3-letter taxonomy codes. Put a short title in
task_subtype (max 200 chars). Optional requesting_agent is a short display
string (max 120 chars) shown on the approve page so the human can see which
agent asked for Proof of Human. Empty or whitespace-only values are omitted.
Do not put PII in any field — the certificate URL is public.
Response:
{
"attestation_id": "uuid",
"approve_url": "https://bridge.trust.sanctifai.com/approve?token=...",
"status": "pending"
}
2. Human step (Chrome)
Tell the human:
Open this HTTPS link in Chrome and confirm with your device passkey:
{approve_url}
The page talks to https://trust.sanctifai.com through this plugin
(presence/start, presence/options, presence/verify). If Trust returns
404 / "no credentials", the page enrolls a passkey on this origin and retries.
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 Changed 48a2eea58f22
- 3d ago First seen · 189 lines · 83 tokens per session scan A 0e16539579a1
proof-of-human is a skill published in the GitHub repository sanctifai/skills (6 stars, last pushed 2d ago), licensed Apache-2.0. It adds 83 tokens to every session and 1,733 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 2026-09-09.
Other skills, from other repositories
ship-it-or-fix-it
Oracle-frozen Builder and independent-Judge convergence cycle. Load ONLY when the operator explicitly sets Governance Dial G2 for the task, or explicitly names this skill or an active work unit already running it. Never auto-activate on task class, such as security, auth, or payments. If a task seems to warrant G2 and…
test-verification
Requires behavioral, failure-path, and durable-seam evidence for tests and review. Use when writing tests, reviewing test coverage, assessing behavioral test quality, or accepting high-risk behavior on test evidence.
portable-adaptive-planning
Use for planning, roadmaps, design, architecture, sequencing, or before consequential work when no current FINAL plan plus GO covers it. Fire before changes to data or schemas, credentials or auth, production or releases, repository or source-home boundaries, irreversible work, or materially costly mistakes. Re-enter…
reasoning-doctrine
The working method for every nontrivial task: frame the objective, ground load-bearing facts, converge on a route, execute without drift, and verify before delivery. Use for analysis, planning, implementation, debugging, document work, reviews, and multi-stage tasks.
write-maintainable-code
Evaluate minimum-sufficient implementation routes and enforce the selected route for a fixed, authorized outcome. Use after outcome, scope, authority, and acceptance evidence are fixed. Do not choose outcomes, set acceptance, adjudicate governance, grade finished work, design tests alone, conduct security audits, or…
gdpr-compliance-checker
Autonomous GDPR compliance auditor that scans a codebase to identify PII collection, storage, and sharing, then produces an article-by-article gap analysis, a pre-filled Data Processing Agreement (DPA), and a ROPA (Record of Processing Activities) starter kit — all exported as downloadable files (.docx recommended).…