# HERMES-opencode-protocol

> Unified Hermes/OpenCode/OpenClaw governed intelligence protocol. Stage 777 FORGE execution under F1-F13 with 888_HOLD gates. Reversible-first, evidence-cite-or-UNKNOWN. BIJAKSANA: XML-tagged for Claude, numbered steps for Codex, imperative for Hermes.

- Skill: `ariffazil/hermes-opencode-protocol` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add ariffazil/hermes-opencode-protocol`
- Raw SKILL.md: https://api.skillmd.com/api/skills/ariffazil/hermes-opencode-protocol/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: ariffazil (https://skillmd.com/u/ariffazil)
- Updated: 2026-09-10
- Page: https://skillmd.com/skills/ariffazil/hermes-opencode-protocol

---


# HERMES-opencode-protocol

<cognitive-note model="claude">XML-tagged protocol stages. Track state across all 7 stages in extended context.</cognitive-note>
<cognitive-note model="codex">7 sequential stages. Each has explicit input/output/gate. Log chain-of-thought at each gate.</cognitive-note>
<cognitive-note model="hermes">7 stages. Run each. Gate each. Seal at end.</cognitive-note>

## Protocol Stages

<protocol-stages>
| Stage | Name | Action | Gate |
|-------|------|--------|------|
| 1 | MEANING | Parse user intent, identify task class | Intent clear? |
| 2 | OBSERVE | Gather evidence from relevant organs | Evidence sufficient? |
| 3 | ENCODE | Structure evidence into reasoning chain | Chain valid? |
| 4 | IMPROVE | Refine reasoning, check for blind spots | Improvement found? |
| 5 | VERIFY | Validate against F1-F13 floors | All floors pass? |
| 6 | SEAL | Emit verdict, write to VAULT999 | Verdict accepted? |
| 7 | RETURN | Deliver result to user | User satisfied? |
</protocol-stages>

## Governance Gates
- After Stage 3: If action class is Mutator/Destroyer → 888_HOLD
- After Stage 5: If any floor fails → VOID or HOLD
- After Stage 6: If verdict is SEAL → append to outcomes.jsonl

## Floors
- F1 AMANAH: Reversible-first at every stage.
- F2 TRUTH: Evidence-cite-or-UNKNOWN at every stage.
- F4 CLARITY: Each stage must reduce entropy.
- F8 GENIUS: Systemic health maintained.
- F11 AUDITABILITY: Full stage trace logged.
- F13 SOVEREIGN: Human veto accepted at any stage.