mirror of
https://github.com/allaunthefox/Research-Stack.git
synced 2026-07-31 03:05:21 +00:00
refactor(agent): break-glass is just a model selection card
OpenCode handles provider routing. The skill is a reference card for which model to switch to. Nothing more. DeepSeek V4 Pro: /usr/bin/bash.27/.10 (55x cheaper than Opus) Claude Opus: 5/5 (baseline)
This commit is contained in:
parent
9c5e5030a5
commit
ec5ade77b5
1 changed files with 20 additions and 80 deletions
|
|
@ -1,90 +1,30 @@
|
|||
# SKILL: break-glass
|
||||
|
||||
## When to Use
|
||||
## Trigger
|
||||
|
||||
Use this skill ONLY when a problem has resisted ALL other approaches:
|
||||
|
||||
1. Local model (current default) — tried and failed
|
||||
2. Single frontier model — tried and failed
|
||||
3. Existing skills and tools — tried and failed
|
||||
|
||||
**Trigger:** `break glass: <problem statement>`
|
||||
`break glass: <problem statement>`
|
||||
|
||||
## What It Does
|
||||
|
||||
Provides model selection guidance based on problem type. Hermes handles the actual model routing.
|
||||
Tells you which model to switch to. OpenCode handles the routing.
|
||||
|
||||
## Model Selection Guide
|
||||
## Model Selection Card
|
||||
|
||||
### Cost Hierarchy (per 1M tokens, input/output)
|
||||
| Model | $/1M in | $/1M out | Best For |
|
||||
|-------|---------|----------|----------|
|
||||
| DeepSeek V4 Flash | $0.07 | $0.28 | Quick checks |
|
||||
| DeepSeek V4 Pro | $0.27 | $1.10 | Math, code (default) |
|
||||
| Gemini 3 Flash | $0.075 | $0.30 | Budget diversity |
|
||||
| Gemini 3.1 Pro | $1.25 | $5.00 | Different perspective |
|
||||
| Kimi K2.6 | $0.50 | $2.00 | Code |
|
||||
| Cohere Command A+ | $2.50 | $10.00 | Structured output |
|
||||
| GLM-5.2 | $1.00 | $4.00 | Different architecture |
|
||||
| GPT-5.5 | $10.00 | $30.00 | Breadth, ops |
|
||||
| Claude Opus 4.8 | $15.00 | $75.00 | Formal proof, synthesis |
|
||||
|
||||
| Model | Input | Output | Provider | Use When |
|
||||
|-------|-------|--------|----------|----------|
|
||||
| **DeepSeek V4 Flash** | $0.07 | $0.28 | DeepSeek | Quick iterations, sanity checks |
|
||||
| **DeepSeek V4 Pro** | $0.27 | $1.10 | DeepSeek | Math reasoning, code, primary workhorse |
|
||||
| **Gemini 3 Flash** | $0.075 | $0.30 | Google | Budget panel diversity |
|
||||
| **Gemini 3.1 Pro** | $1.25 | $5.00 | Google | Different thinking style |
|
||||
| **Kimi K2.6** | $0.50 | $2.00 | Moonshot | Code, different training |
|
||||
| **Cohere Command A+** | $2.50 | $10.00 | Cohere | Structured output, tool-use |
|
||||
| **GLM-5.2** | $1.00 | $4.00 | Zhipu | Architectural diversity (753B) |
|
||||
| **GPT-5.5** | $10.00 | $30.00 | OpenAI | Breadth, ops knowledge |
|
||||
| **Claude Opus 4.8** | $15.00 | $75.00 | Anthropic | Formal proof, synthesis |
|
||||
## Escalation
|
||||
|
||||
### Problem Type → Model Recommendation
|
||||
|
||||
| Problem Type | First Try | If Stuck | Last Resort |
|
||||
|-------------|-----------|----------|-------------|
|
||||
| **lean_proof** | DeepSeek V4 Pro | + Claude Opus | + GPT-5.5 + Cohere |
|
||||
| **code_bug** | DeepSeek V4 Pro | + GPT-5.5 | + Claude Opus |
|
||||
| **architecture** | Claude Opus | + DeepSeek V4 Pro | + GPT-5.5 + Cohere |
|
||||
| **math_novel** | DeepSeek V4 Pro | + Claude Opus | + Gemini 3.1 Pro |
|
||||
| **infra_debug** | GPT-5.5 | + DeepSeek V4 Pro | + Claude Opus |
|
||||
| **hard_wall** | DeepSeek V4 Pro | + Claude Opus + GPT-5.5 | All 8 models |
|
||||
|
||||
### The "All 8 Models" Panel
|
||||
|
||||
For truly stuck problems, switch between all 8 models in sequence:
|
||||
1. DeepSeek V4 Pro ($0.27/$1.10) — math reasoning
|
||||
2. Claude Opus ($15/$75) — formal proof, synthesis
|
||||
3. GPT-5.5 ($10/$30) — breadth, ops
|
||||
4. Cohere Command A+ ($2.50/$10) — structured output
|
||||
5. Gemini 3.1 Pro ($1.25/$5) — different perspective
|
||||
6. GLM-5.2 ($1/$4) — different architecture
|
||||
7. Kimi K2.6 ($0.50/$2) — code
|
||||
8. DeepSeek V4 Flash ($0.07/$0.28) — fast iteration
|
||||
|
||||
### Cost-Optimal Strategy
|
||||
|
||||
**Default:** DeepSeek V4 Pro — 55x cheaper than Opus, nearly as good at math/code.
|
||||
|
||||
**Escalation path:**
|
||||
1. Start with DeepSeek V4 Pro ($0.27/$1.10)
|
||||
2. If stuck, add Claude Opus for synthesis ($15/$75)
|
||||
3. If still stuck, add GPT-5.5 for breadth ($10/$30)
|
||||
4. If truly stuck, use all 8 models sequentially
|
||||
|
||||
**Budget per problem:**
|
||||
- Routine: $0.10-0.50 (DeepSeek only)
|
||||
- Stuck: $1.00-3.00 (DeepSeek + Opus)
|
||||
- Hard wall: $5.00-10.00 (all 8 models)
|
||||
- Break glass: $15.00+ (all 8 + multiple rounds)
|
||||
|
||||
## How to Use
|
||||
|
||||
```
|
||||
break glass: cleanMerge_preservesGap sorry — List.zip/filter/all terms too large for simp
|
||||
```
|
||||
|
||||
The agent will:
|
||||
1. Classify the problem type
|
||||
2. Recommend which model to switch to
|
||||
3. Provide context to pass to that model
|
||||
4. Track cost
|
||||
|
||||
## Files
|
||||
|
||||
| File | Purpose |
|
||||
|------|---------|
|
||||
| `.opencode/skills/break-glass/SKILL.md` | This file |
|
||||
| `6-Documentation/docs/specs/break_glass_problem_statement.md` | Problem statement |
|
||||
| `6-Documentation/docs/specs/break_glass_mcp_agent.md` | Design spec (historical) |
|
||||
1. **DeepSeek V4 Pro** — try first ($0.27/$1.10, 55x cheaper than Opus)
|
||||
2. **Claude Opus** — if DeepSeek stuck ($15/$75)
|
||||
3. **GPT-5.5** — if Opus stuck ($10/$30)
|
||||
4. **All 8** — hard wall ($5-10 total)
|
||||
|
|
|
|||
Loading…
Add table
Reference in a new issue