SyslogSolution/prose-contracts
7
0
Fork 0
Code Issues 2 Pull Requests Actions Packages Projects Releases Wiki Activity

Item 5: Hermes context-detection fix — Rule 14

Browse Source 
- hermes-config-template.prose.md: Add Rule 14 — Hermes reads max_model_tokens (128K), NOT max_input_tokens (64K)
- Warning: Using max_input_tokens for Hermes agents causes premature context loss
- Crewmates use max_input_tokens (64K cap), Abiba/Hermes use max_model_tokens (128K)
This commit is contained in:
root
2026-08-20 07:53:58 +00:00
parent b7778c287b
commit 819f2f400d
1 changed files with 11 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
hermes-config-template.prose.md
+11
View File
@@ -338,6 +338,17 @@ curl -s -o /dev/null -w 'key_health: %{http_code}\n' -H "Authorization: Bearer $
### Rule 13: API Key Injection — Two Patterns (UPDATED 2026-07-16, WAL #1300)
### Rule 14: Hermes Context Detection Uses `max_model_tokens`, NOT `max_input_tokens`
**CRITICAL**: Hermes context detection reads `max_model_tokens` (128K), NOT `max_input_tokens` (64K cap).
- **Abiba and Hermes agents**: `max_model_tokens: 131072` (128K) — unlimited context
- **Crewmates (ops, tune, verify, auth-keys, build)**: `max_input_tokens: 64000` (64K) — capped
- If you see `max_input_tokens: 64000` in an Abiba/Hermes config, that's a mistake
- Using `max_input_tokens` for Hermes agents causes premature context loss
- Check: `grep -n 'max_model_tokens\|max_input_tokens' ~/.hermes/config.yaml`
- Expected output: `max_model_tokens: 131072` (not max_input_tokens)
Agents inject `LITELLM_API_KEY` via ONE of two mechanisms. Both are valid; the contract
requirement is that the key is a **valid LiteLLM virtual key** (HTTP 200 on /v1/models).