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 rrpauls/hermes-esra --skill loop-auditorgit clone --depth 1 https://github.com/rrpauls/hermes-esraWrote 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/rrpauls/hermes-esra/loop-auditor)<a href="https://agentmods.dev/skills/rrpauls/hermes-esra/loop-auditor"><img src="https://agentmods.dev/badge/skills/rrpauls/hermes-esra/loop-auditor/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/rrpauls/hermes-esra/loop-auditor"><img src="https://agentmods.dev/badge/skills/rrpauls/hermes-esra/loop-auditor.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 | $0.00082 | $0.01067 |
| Opus 5.5 | $0.00033 | $0.00427 |
| Sonnet 5.5 | $0.00016 | $0.00213 |
| Haiku 4.5 | $0.00008 | $0.00107 |
Grade A, and why
loop-auditor 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 1mo 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 — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Loop Auditor
Role
You are a meta-auditor of the agent's self-evolution system. Your job is to periodically examine the entire Self-Evolution Loop, assess its health, identify structural weaknesses, and recommend improvements to how the agent develops itself.
When This Skill Activates
Use this skill when:
- Conducting a periodic review of the development system.
- The agent suspects the improvement process itself has become inefficient or misaligned.
- Major changes have occurred in capabilities or environment.
- Explicitly requested to audit the evolution loop.
- New triggers for Hermes context:
- After several cycles of
hermes-evolution-orchestrator - When observed skill-creation or improvement workflows seem slow, shallow, or misaligned
- To audit the integration quality between recorded Hermes workflows and our meta-skills (ooda-framework, orchestrator, self-improver, etc.)
- Periodically (recommended: every 5–10 significant evolution cycles)
- After several cycles of
Core Audit Process
0. Audit Hermes + Orchestrator Integration (Hermes-specific)
- How effectively does
hermes-evolution-orchestratorconnect observed Hermes workflows with our meta-skills? - Is the orchestrator being invoked deliberately after qualifying complex tasks or skill changes?
- Are the meta-skills (especially
ooda-framework,self-improver,mental-model-updater) actually improving the quality and depth of Hermes-generated skills? - Identify friction points between host behavior and our deliberate evolutionary layer without assuming unobserved native automation.
- Check whether Orientation phase (in OODA) is sufficiently leveraged inside the orchestrator.
1. Map Current State of the Loop
Identify active skills and interaction patterns from recorded evidence (including Hermes workflows + orchestrator).
- Game-Theoretic Analysis
- Analyze interactions between skills as a repeated game.
- Check for stable but suboptimal equilibria (Nash vs Pareto).
- Evaluate reputation, cooperation, and defection patterns between skills.
- Assess whether the "rules of interaction" encourage good outcomes.
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.
- 1mo ago Changed · +1 lines 8028f0bdc0e6
- 1mo ago First seen · 86 lines · 82 tokens per session scan A 2f4a9a40b3fc
loop-auditor is a skill published in the GitHub repository rrpauls/hermes-esra (1 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 82 tokens to every session and 1,067 once invoked, about $0.0003 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.
Other skills, from other repositories
local-knowledge-router
Route local operational questions to indexed skills, scripts, runbooks, cron jobs, MCP servers, and docs before guessing paths or doing broad file search.
svix-sending-webhooks
Everything for working with Svix webhooks: first-time setup (API key, SDK install, first message), Dispatch (sending webhooks to your customers), Ingest (receiving third-party webhooks), Applications, Channels, customer UIDs, idempotency, App Portal embedding, operational webhooks, the Svix CLI, and — only when the…
receiving-webhooks
General guidelines for building a robust webhook receiver/handler: verifying signatures, raw-body access, replay protection, async processing, retries and endpoint auto-disabling. Use whenever you write, review, or debug a handler that consumes incoming webhooks from any provider.
skill-auto-maintain
Automated lifecycle management for non-bundled Hermes skills. Scans all skill directories, migrates orphans to userskills/, registers in userskills.json, detects merge candidates via multi-dimensional similarity, and optimises SKILL.md format. No SOUL framework dependency — works on any standard Hermes Agent…
local-knowledge-router
Route local operational questions to indexed skills, scripts, runbooks, cron jobs, MCP servers, and docs before guessing paths or doing broad file search.
orbit-notion
Open Orbit briefing skill — selected by the Orbit pipeline when Notion is the user's only connected connector, or when the user explicitly scopes their daily digest to Notion. Pulls the past 24 hours of document edits, comments, mentions, and database row changes from the user's authenticated Notion connection and…