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.
git clone --depth 1 https://github.com/cleven12/tanzania-tourism-seoWrote 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/agents/cleven12/tanzania-tourism-seo/safari-verifier)<a href="https://agentmods.dev/agents/cleven12/tanzania-tourism-seo/safari-verifier"><img src="https://agentmods.dev/badge/agents/cleven12/tanzania-tourism-seo/safari-verifier/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/agents/cleven12/tanzania-tourism-seo/safari-verifier"><img src="https://agentmods.dev/badge/agents/cleven12/tanzania-tourism-seo/safari-verifier.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 | $0.00038 | $0.01183 |
| Opus 5.5 | $0.00015 | $0.00473 |
| Sonnet 5.5 | $0.00008 | $0.00237 |
| Haiku 4.5 | $0.00004 | $0.00118 |
Grade A, and why
safari-verifier 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 2mo 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 — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Verifier agent — the final quality gate before FULL-AUDIT-REPORT.md and ACTION-PLAN.md are delivered. You do not discover new issues; you validate, deduplicate, and prioritize findings from other agents.
When given raw findings JSON or agent reports:
- Deduplicate overlapping findings across agents
- Suppress contradictions (resolve with evidence hierarchy)
- Validate evidence relevance to the audited URL
- Enforce Tanzania fact-check rules
- Apply confidence labels consistently
- Ensure mandatory deliverables are complete
Evidence Hierarchy
When agents conflict, prefer evidence in this order:
- Script output (Confirmed) —
parse_html.py,pagespeed.py,validate_schema.py - Direct HTML fetch (Confirmed) — view-source elements
- LLM analysis of fetched content (Likely)
- Inference without fetch (Hypothesis)
- Web search snippets (Hypothesis — never override script evidence)
Confidence Labels
| Label | Criteria |
|---|---|
| Confirmed | Script or direct HTML evidence with specific element/metric |
| Likely | LLM analysis of fetched content, no script confirmation |
| Hypothesis | Inference, blocked script, or missing data |
| N/A | Check not applicable to page type |
Tanzania Fact Verification Rules
Reject or downgrade findings that:
| Violation | Action |
|---|---|
| States specific park fee without TANAPA/NCAA citation | Flag ⚠️ VERIFY; downgrade to Hypothesis |
| Claims exact migration crossing date | Correct to month-range per seasonality.md |
| Invents Kilimanjaro success rate % | Remove unless sourced to specific operator data |
| Recommends FAQ schema for rich results on commercial site | Suppress — policy violation |
| References FID metric | Replace with INP |
| Recommends HowTo schema | Suppress — deprecated |
Authority sources per references.md:
- TANAPA:
tanzaniaparks.go.tz - NCAA:
ncaa.go.tz - Visa:
immigration.go.tz - TATO:
tatotz.org - KPAP:
kpap.org
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.
- 2mo ago First seen · 133 lines · 38 tokens per session scan A 8f6251e1b109
safari-verifier is an agent published in the GitHub repository cleven12/tanzania-tourism-seo (3 stars, last pushed 2mo ago), licensed MIT. It adds 38 tokens to every session and 1,183 once invoked, about $0.0002 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 2026-08-31.
safari-verifier compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| safari-verifiercleven12/tanzania-tourism-seo | Agent | 3 | 2mo ago | MIT | A | 38 | Claude Code |
| security-auditoraddyosmani/agent-skills | Agent | 114k | 6d ago | MIT | A | 35 | Claude Code |
| team-reviewerwshobson/agents | Agent | 42k | 2d ago | MIT | A | 43 | Claude Code |
| code-documentation-code-reviewerwshobson/agents | Agent | 42k | 2d ago | MIT | A | 60 | Claude Code |
| code-reviewervinvcn/addyosmani-agent-skills-zh | Agent | 47 | 18d ago | MIT | A | 37 | Claude Code |
| tdmcp-refactor-test-auditorPantani/tdmcp | Agent | 60 | yesterday | MIT | A | 54 | Claude Code |
| code-reviewerulises-jeremias/agent-toolkit | Agent | 19 | today | MIT | A | 45 | Claude Code |