Skip to content

Worktrunk is a CLI for Git worktree management, designed for parallel AI agent workflows

docs: add opencode and condense other LLM commit tools (#1295)

worktrunk-botb49ab2bbc028
diff --git a/docs/content/llm-commits.md b/docs/content/llm-commits.md
index 480d22091..25f2d6463 100644
--- a/docs/content/llm-commits.md
+++ b/docs/content/llm-commits.md
@@ -29,33 +29,28 @@ command = "CLAUDECODE= MAX_THINKING_TOKENS=0 claude -p --model=haiku --tools=''
`CLAUDECODE=` unsets the nesting guard so `claude -p` works from within a Claude Code session. The other flags disable tools, skills, settings, and system prompt for fast text-only output. See [Claude Code docs](https://docs.anthropic.com/en/docs/build-with-claude/claude-code) for installation.
-### llm
+### Codex
```toml
[commit.generation]
-command = "llm -m claude-haiku-4.5"
+command = "codex exec -m gpt-5.1-codex-mini -c model_reasoning_effort='low' --sandbox=read-only --json - | jq -sr '[.[] | select(.item.type? == \"agent_message\")] | last.item.text'"
```
-Install with `uv tool install llm llm-anthropic && llm keys set anthropic`. See [llm docs](https://llm.datasette.io/).
+Uses the fast mini model with low reasoning effort. Requires `jq` for JSON parsing. See [Codex CLI docs](https://developers.openai.com/codex/cli/).
-### aichat
+### Other tools
```toml
-[commit.generation]
-command = "aichat -m claude:claude-haiku-4.5"
-```
-
-See [aichat docs](https://github.com/sigoden/aichat).
+# opencode — use a fast model variant
+command = "opencode run -m anthropic/claude-haiku-4.5 --variant fast"
-### Codex
+# llm
+command = "llm -m claude-haiku-4.5"
-```toml
-[commit.generation]
-command = "codex exec -m gpt-5.1-codex-mini -c model_reasoning_effort='low' --sandbox=read-only --json - | jq -sr '[.[] | select(.item.type? == \"agent_message\")] | last.item.text'"
+# aichat
+command = "aichat -m claude:claude-haiku-4.5"
```
-Uses the fast mini model with low reasoning effort. Requires `jq` for JSON parsing. See [Codex CLI docs](https://developers.openai.com/codex/cli/).
-
## How it works
When worktrunk needs a commit message, it builds a prompt from a template and pipes it to the configured command via shell (`sh -c`). Environment variables can be set inline in the command string.
diff --git a/skills/worktrunk/reference/llm-commits.md b/skills/worktrunk/reference/llm-commits.md
index 84c4f68b4..cb1cbb4b1 100644
--- a/skills/worktrunk/reference/llm-commits.md
+++ b/skills/worktrunk/reference/llm-commits.md
@@ -15,33 +15,28 @@ command = "CLAUDECODE= MAX_THINKING_TOKENS=0 claude -p --model=haiku --tools=''
`CLAUDECODE=` unsets the nesting guard so `claude -p` works from within a Claude Code session. The other flags disable tools, skills, settings, and system prompt for fast text-only output. See [Claude Code docs](https://docs.anthropic.com/en/docs/build-with-claude/claude-code) for installation.
-### llm
+### Codex
```toml
[commit.generation]
-command = "llm -m claude-haiku-4.5"
+command = "codex exec -m gpt-5.1-codex-mini -c model_reasoning_effort='low' --sandbox=read-only --json - | jq -sr '[.[] | select(.item.type? == \"agent_message\")] | last.item.text'"
```
-Install with `uv tool install llm llm-anthropic && llm keys set anthropic`. See [llm docs](https://llm.datasette.io/).
+Uses the fast mini model with low reasoning effort. Requires `jq` for JSON parsing. See [Codex CLI docs](https://developers.openai.com/codex/cli/).
-### aichat
+### Other tools
```toml
-[commit.generation]
-command = "aichat -m claude:claude-haiku-4.5"
-```
-
-See [aichat docs](https://github.com/sigoden/aichat).
+# opencode — use a fast model variant
+command = "opencode run -m anthropic/claude-haiku-4.5 --variant fast"
-### Codex
+# llm
+command = "llm -m claude-haiku-4.5"
-```toml
-[commit.generation]
-command = "codex exec -m gpt-5.1-codex-mini -c model_reasoning_effort='low' --sandbox=read-only --json - | jq -sr '[.[] | select(.item.type? == \"agent_message\")] | last.item.text'"
+# aichat
+command = "aichat -m claude:claude-haiku-4.5"
```
-Uses the fast mini model with low reasoning effort. Requires `jq` for JSON parsing. See [Codex CLI docs](https://developers.openai.com/codex/cli/).
-
## How it works
When worktrunk needs a commit message, it builds a prompt from a template and pipes it to the configured command via shell (`sh -c`). Environment variables can be set inline in the command string.