Skip to content

graff — a fast agentic coding harness in Zig: multi-provider, MCP, workflows, DGM evolution loop, TS/Python SDKs

Python

Prepare v0.302.62 release

justrach9c2c1349577b
diff --git a/CHANGELOG.md b/CHANGELOG.md
index eaa5310e..500926a8 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -10,6 +10,11 @@ The release workflow uses a tag's section here as its release notes (a
hand-written `docs/releases/<tag>.md` wins if present), so keeping this file
current is part of cutting a release.
+## v0.302.62
+
+- `rlm`'s `llm_query` works on Claude Agent. Each sub-query runs on its own short-lived, tools-off `claude` process, the way titles and recaps do, so the chat's `claude` process and its prompt cache carry on untouched. Before, it answered "llm_query is not available on Claude Agent" (#1777, ADR 0309).
+- Retains all changes from v0.302.61.
+
## v0.302.61
- Claude Agent: Claude models through your own Claude Code (`claude`) and its login. Sign in with `graff login claude-agent` (or `/login claude-agent`); graff checks `claude auth status`, never reads Claude Code's credentials, and runs your installed `claude` with graff's tools, prompt, history and loop. Calls are flat-rate on a Claude plan (#1773, ADR 0309).
diff --git a/docs/releases/v0.302.62.md b/docs/releases/v0.302.62.md
new file mode 100644
index 00000000..ab9a0794
--- /dev/null
+++ b/docs/releases/v0.302.62.md
@@ -0,0 +1,7 @@
+# v0.302.62
+
+`rlm`'s `llm_query` works on Claude Agent.
+
+- **Sub-queries run on Claude Agent.** Inside `rlm`, `llm_query` used to answer "llm_query is not available on Claude Agent" because that provider has no HTTP endpoint. Each sub-query now runs on its own short-lived `claude` process with no tools, the way titles and recaps already do, and the text comes back as usual. The chat's own `claude` process and its prompt cache are never touched (#1777, ADR 0309).
+
+Includes all changes from v0.302.61.
diff --git a/scripts/eval/tier1-manifest.json b/scripts/eval/tier1-manifest.json
index 14d29cbc..1aa73bcd 100644
--- a/scripts/eval/tier1-manifest.json
+++ b/scripts/eval/tier1-manifest.json
@@ -14,7 +14,7 @@
"src/main.zig",
"src/repl.zig"
],
- "test_count_baseline": 3575,
+ "test_count_baseline": 3576,
"test_count_slack": 25,
"required_invariants": [
{
diff --git a/src/cli_text_source.zig b/src/cli_text_source.zig
index b1aec03a..786bc7f5 100644
--- a/src/cli_text_source.zig
+++ b/src/cli_text_source.zig
@@ -5,6 +5,9 @@
pub const changelog_text =
\\What's new
\\──────────
+ \\0.302.62
+ \\ • rlm's llm_query works on Claude Agent, on a throwaway claude process
+ \\
\\0.302.61
\\ • Claude Agent: Claude models through your own Claude Code login (graff login claude-agent)
\\ • subagent requests keep agent_output in lean mode, so the turn waits for its workers