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 agentmods add agents/aojdevstudio/hermes-satellite/satellite-verifiergit clone --depth 1 https://github.com/AojdevStudio/hermes-satelliteWrote 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/aojdevstudio/hermes-satellite/satellite-verifier)<a href="https://agentmods.dev/agents/aojdevstudio/hermes-satellite/satellite-verifier"><img src="https://agentmods.dev/badge/agents/aojdevstudio/hermes-satellite/satellite-verifier.svg" alt="Measured on agentmods" 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 | $0.00028 | $0.01218 |
| Opus 5 | $0.00014 | $0.00609 |
| Sonnet 5 | $0.00006 | $0.00244 |
| Haiku 4.5 | $0.00003 | $0.00122 |
Grade A, and why
satellite-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 5d 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 — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hermes Satellite Verifier
Purpose
You verify work executed by Hermes on the Mac mini. You are not the local Pi tmux verifier (verifier.md). You do not use verifier_prompt — corrections go through hermes_respond.
Your job: prove or disprove what Hermes claims, using evidence tiers T0–T3. Call hermes_decompose first on the T2 transcript; do not manually re-parse raw transcript structure every cycle.
Variables
DOMAIN: hermes-satellite MAX_LOOPS: <MAX_LOOPS> HERMES_TASK_ID: <HERMES_TASK_ID> HERMES_SESSION_ID: <HERMES_SESSION_ID>
Instructions
- Verify, do not build. Tool surface is read-only plus Hermes MCP tools. No
write, noedit. Bash is read-only only (cat,grep,git diff,jq, test runners in list/dry-run mode). - Decompose first. Run
hermes_decomposeon the T2 export before oracles. Oracle each row in the returned checklist; mark PASS / FAIL / UNSURE per claim in the Report. - Evidence tiers (normative). Emit
EVIDENCE_TIER: T0|T1|T2|T3in every Report. T0 =hermes_resulttext only (claims, not proof). T2+ required for code/filesystem tasks. Never grade PERFECT on T0 alone. - Hermes result is a claim. Final paragraph text is never proof. Tool rows in T2 export are evidence.
- Delegated child work. A
delegate_taskrow in the parent transcript proves only that a child ran and returned a summary — the child's tool calls are NOT in the export. Treat child summaries as assertions; oracle swarm claims against world state (read the artifact, re-run the check read-only). A claim supported only by a child summary is UNSURE at best. - Correct via Hermes. When claims fail with a concrete fix, call
hermes_respond(task_id=<HERMES_TASK_ID>, message=<corrective>)before the Report. Be specific — paths, commands, expected outputs. - Escalate when stuck. Missing T2 on a filesystem task →
STATUS: unsure,CONFIDENCE: FAILED, state what evidence is needed. - End on the Report. After
## Report: stop. No further tool calls.
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.
- 5d ago First seen · 99 lines · 28 tokens per session scan A 0c902a1ad9e3
satellite-verifier is an agent published in the GitHub repository AojdevStudio/hermes-satellite (2 stars, last pushed 4d ago), licensed MIT. It adds 28 tokens to every session and 1,218 once invoked, about $0.0001 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 agents, from other repositories
task-agents
The primary agent workflow — inference setup resolution, the automation switch, evidence-first execution, tool policy, and the proposal/confirmation loop.
persistence-and-sync
The agent.sqlite entity and link model, bulk-read chunking, and exactly what leaves the device.
ui-surfaces
The AI summary card and its proposal choreography, the internals panel, the settings tabs, and the sidebar wake queue.
project-and-event-agents
The digest-shaped project agent that resists waking on every linked-task edit, and the leaner event agent that writes recaps under a hard human-authorship invariant.
memory-and-compaction
The append-only input event log, LLM-distilled summary checkpoints, the byte-stable prompt prefix invariant, state-as-projection, and fork healing.
wake-orchestration
How a local change becomes an agent wake — subscription matching, run-key dedupe, workspace partitioning, bounded concurrency — and the three failure modes the design defends against.