Skip to content

ENG-5731 docs: CLI-first agent integrations (pioneer integrate) - #61

Open
samrathreddy wants to merge 2 commits into
fastino-ai:mainfrom
samrathreddy:docs/cli-integrate-agents
Open

ENG-5731 docs: CLI-first agent integrations (pioneer integrate)#61
samrathreddy wants to merge 2 commits into
fastino-ai:mainfrom
samrathreddy:docs/cli-integrate-agents

Conversation

@samrathreddy

Copy link
Copy Markdown
Member

ENG-5731

Rewrites the Mintlify docs for the five coding-agent integrations to the CLI-first pioneer integrate flow, replacing the long manual setup steps (jq catalog builders, gateway wiring, inline hook scripts).

Pages: OpenCode, OpenClaw, Hermes, Claude Code, Codex.

Each page now:

  • Install + pioneer auth login (drop in your Pioneer API key) → pioneer integrate <agent> → pick a model.
  • A ## Flags section documenting the two flag namespaces truthfully:
    • pioneer integrate --help = the Pioneer CLI's options; <agent> --help (run directly) = the agent's own.
    • Pioneer flags (--model, --level) go before the agent name; anything after the agent name is forwarded to the agent. Codex/Claude Code show a forwarded dangerous-bypass flag as the example.
  • No pioneer/auto or Code Router references; model example uses kimi-k3.
  • Restore/undo documented per agent (OpenClaw, OpenCode, Hermes are invasive; Codex, Claude Code are non-invasive).

Diff: ~211 insertions, ~1309 deletions across the 5 pages.

Linear: https://linear.app/personal-fastino/issue/ENG-5731

…I flow

Rewrite the following pages around the pioneer integrate CLI flow:
- opencode.mdx
- openclaw.mdx
- hermes.mdx
- claude-code.mdx
- codex.mdx
Refine the pioneer integrate rewrite for the 5 agent pages:
- Add a note that pioneer auth login prompts you to drop in your API key.
- Remove pioneer/auto and Code Router references; use kimi-k3 as the model example.
- Add a dedicated ## Flags section (after the steps) documenting the two flag
  namespaces truthfully: pioneer integrate --help vs running the agent's own
  --help directly, and that Pioneer flags (--model/--level) go before the agent
  name while anything after the agent name is forwarded to it.
- Codex and Claude Code show a forwarded dangerous-bypass flag as the example.
- Fix --model ordering to place the flag before the agent name.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant