bazhand
The quite useful tool is to use /opusplan along with /codex:rescue (https://github.com/openai/codex-plugin-cc) means you get quite a strongly reviewed plan using native claude + codex without having to implement the mostly useless trust-me-bro plugins and other bs.
The fix I (well Codex actually) landed on was toolset tiers (minimal/authoring/experimental) controlled by env var, plus phase-gating, now tools are registered but ~80% are "not connected" until you call _connect. The effective listed surface stays pretty small.
Lazy loading basically, not a new concept for people here.
[1] https://skills.sh/vercel-labs/agent-skills/web-design-guidel... [2] https://github.com/vercel-labs/agent-skills/blob/main/skills...
All of these SKILLS.md/AGENTS.md/COMMANDS.md are just simple prompts, maybe even some with context links.
And quite dangerous.
This is exactly why (when OpenCode and Charm/Crush started diverging) Charm chose not to support “use your Claude subscription” auth and went in a different direction (BYOK / multi-provider / etc). They didn’t want to build a product on top of a fragile, unofficial auth path.
And I think there’s a privacy/policy reason tightening this now too: the recent Claude Code update (2.1-ish) pops a “Help improve Claude” prompt in the terminal. If you turn that ON, retention jumps from 30 days to up to 5 years for new/resumed chats/coding sessions (and data can be used for model improvement). If you keep it OFF, you stay on the default 30-day retention. You can also delete data anytime in settings. That consent + retention toggle is hard to enforce cleanly if you’re not in an official client flow, so it makes sense they’re drawing a harder line.
I guess what I'm asking is, if there is a memory block limit, is that an issue for self learning over time. Claude as you know just straight up ignores CLAUDE.md and doesnt self-improve it.
Subagents using MCP would solve quite a lot of the context preparation required for a successful session. I'm using GLM-4.6 vision/web search MCP in claude code to relative success as the subagents can access these via API (not true MCP behaviour but close enough)
Add these settings to your `.claude/settings.json`:
```json
{
"env": {
"ANTHROPIC_CUSTOM_HEADERS": {"anthropic-beta":
"context-1m-2025-08-07"},
"ANTHROPIC_MODEL": "claude-sonnet-4-20250514",
"CLAUDE_CODE_MAX_OUTPUT_TOKENS": 8192
}
}
```Opencode allows auth via Claude Max, which is a huge plus over requiring API (ANTHROPIC_API_KEY)
Mostly I'm disappointed that the higher tiers are just usage tiers rather than features.
You can’t sign in because this app sent an invalid request. You can try again later or contact the developer about this issue. Learn more about this error If you are a developer of this app, see error details. Error 400: redirect_uri_mismatch
- you run advertising and lead gen on behalf of clients using their CRM (their domain email/outreach)
- and require Stripe for attribution, and to remit payment from client to you
That's most likely triggering Stripe's TOS.
For something so core to your business, did you have a discussion with Stripe if this was allowed?
Some observations - clicking browser back button loses inputs - ukraine not available? - needs date/month selectors to plan ahead
The creator has quite a soothing demeanour.