skills.rest
AtlasOmnia
AtlasOmniaCOMMUNITY
Explore all skills by this author →
Website𝕏@AtlasOmniaGitHub

hermes-config-editing

Edit Hermes Agent config.yaml values using section-aware patterns that bypass security guards and CLI persistence quirks.

115|9|Updated Aug 5, 2026
Search Tags:#editing#troubleshooting#cli#configuration#yaml#compression#hermes
One-click install
npx skills add https://github.com/AtlasOmnia/donna-starter --skill hermes-config-editing-atlasomnia
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: hermes-config-editing
Source: https://github.com/AtlasOmnia/donna-starter/tree/main/skills/autonomous-ai-agents/hermes-config-editing
Command: npx skills add https://github.com/AtlasOmnia/donna-starter --skill hermes-config-editing-atlasomnia

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, and includes scripts (resource) and references (resource) components.

What problem does it solve? Editing Hermes Agent configuration is unreliable: direct file writes to config.yaml are blocked by a security guard, and hermes config set often fails to persist nested YAML keys. This Skill provides verified patterns for inspecting, editing, and validating configuration values without corrupting file formatting or touching unrelated sections. ## Core Features & Use Cases - Section-aware YAML editing: Python patterns that modify only the target section (e.g. compression:) while preserving comments, ordering, and unrelated keys. - Compression and auxiliary provider configuration: Set compression thresholds, target ratios, and switch auxiliary task providers (web_extract, compression, tts, etc.) with batched CLI commands and readback verification. - Troubleshooting guides: Diagnose context-length auto-detection mismatches with LM Studio, truncated API keys, and Mnemosyne sleep versus legacy memory optimization confusion. - Use Case: Your Hermes session reports 131k context but LM Studio has a 98k GGUF loaded. Use this Skill to identify the family-fallback bug in model_metadata.py and either apply the native API patch or set model.context_length explicitly, then verify via the model info endpoint. ## Quick Start Ask the agent to inspect the current compression settings in ~/.hermes/config.yaml and change the threshold to 0.85 using a section-aware edit, then verify the YAML still parses.

Frequently Asked Questions about hermes-config-editing

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I edit Hermes config.yaml when direct writes are blocked?▼

Direct file edits to config.yaml are blocked by a security guard, so use the Hermes venv Python to inspect values with yaml.safe_load, then apply section-bounded text edits that only modify the target section. Always re-validate the YAML after writing.

Why does hermes config set not persist my changes?▼

The CLI setter often fails silently for nested YAML paths like compression.threshold. Verify with a readback via hermes config show or yaml.safe_load, and fall back to a section-aware Python edit for nested structures.

How do I fix wrong context length detection with LM Studio?▼

The mismatch comes from fallback logic in agent/model_metadata.py that cannot match LM Studio aliases and defaults to 131072 for qwen-family names. Either apply the native API patch for reverse-slug matching or set model.context_length explicitly to your GGUF's loaded context.

What is the difference between Mnemosyne sleep and memory optimization cron?▼

Mnemosyne sleep consolidates working memories into episodic layers via mnemosyne_sleep, while legacy memory optimization manages MEMORY.md and USER.md files. A disabled optimization cron does not block Mnemosyne consolidation; verify with mnemosyne_sleep(dry_run=true).

How do I switch auxiliary providers for compression and web_extract?▼

Batch hermes config set commands for each auxiliary task key (provider, model, base_url, api_key), then grep config.yaml for leftover references to the old provider. Keep auxiliary.vision separate since it typically uses a local vision model.

Related Skills

Explore More Software Engineering Skills→
GitnappGitnapp

hermes-profile-config

Migrate Hermes Agent profile configurations across config.yaml and .env files.

Community
Advanced
vTRKAvTRKA

audit

Audit health-check artifacts and runtime state for staleness and quality gaps.

Community
Advanced
AtlasOmniaAtlasOmnia
115

hermes-mnemosyne

Configure, troubleshoot, and operate the Mnemosyne memory provider for Hermes Agent.

Community
Advanced
ascorbicascorbic
120

onboarding

Automate Macrodata initial setup by creating identity, profile, and workspace state files.

Community
Advanced