golem15/msd-core
4
0
Fork 0
Code Issues 1 Pull Requests Actions 1 Packages Projects Releases Wiki Activity

feat(#2481): add a negotiated effortSurface axis and wire invocation-time effort

Browse Source 
ADR-1239 gains a ninth negotiated axis, effortSurface (argv | none), declaring how
a host accepts reasoning effort. ADR-443 is amended in the same change because its
recorded deferral is what the axis resolves: its Unblock condition offered paths
(a) and (b) and stated the choice was 'a maintainer call this file records but does
not make'. Path (a) is selected and satisfied here.

Before this, effort reached a runtime only through install-time channels
(EFFORT_RENDERING's frontmatter/api), so reviewer CLIs spawned as subprocesses
silently inherited whatever effort sat in the user's own global CLI config. The
review lane now resolves one universal effort through the ADR-443 cascade and
renders it per host through the negotiated descriptor.

Every per-host value is documentation-sourced, never inferred:
- claude   argv  -- verified via 'claude --help' (--effort <level>)
- opencode argv  -- verified via 'opencode run --help' (--variant)
- codex    argv  -- codex-rs/exec/src/cli.rs: model_reasoning_effort is NOT a CLI
                    flag (config.toml key only), so the global -c override is the
                    only argv route
- 15 hosts undocumented -- their docs state no reasoning setting; the sentinel
                    fails closed rather than inheriting a profile baseline

No config-file vocabulary member: the only host that ever had one (Gemini CLI's
thinkingConfig) was removed as a sunset runtime by 8f2ebbe9b (#1928, PR #1996),
and neither Antigravity CLI nor ZCode documents a reasoning setting.

Closes #2481

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Tom Boucher
2026-07-21 09:12:32 -04:00
parent c19d3d7bda
commit 09b535ac00
51 changed files with 863 additions and 100 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
gsd-core/workflows/review.md
View File

@@ -250,6 +250,12 @@ OPENCODE_MODEL=$(gsd_run query config-get review.models.opencode 2>/dev/null | j
# pinned model that 404s server-side); otherwise agy uses its persisted default.
AGY_MODEL=$(gsd_run query config-get review.models.agy 2>/dev/null | jq -r '.' 2>/dev/null || true)
# Reasoning effort per reviewer (#2481). Empty unless the host's effortSurface
# axis is `argv`. Pass --attempt N to walk ADR-443's escalation ladder.
CLAUDE_EFFORT_ARGS=$(gsd_run query resolve-execution gsd-plan-checker --host claude 2>/dev/null | jq -r '.effort_argv_string // ""' 2>/dev/null || true)
CODEX_EFFORT_ARGS=$(gsd_run query resolve-execution gsd-plan-checker --host codex 2>/dev/null | jq -r '.effort_argv_string // ""' 2>/dev/null || true)
OPENCODE_EFFORT_ARGS=$(gsd_run query resolve-execution gsd-plan-checker --host opencode 2>/dev/null | jq -r '.effort_argv_string // ""' 2>/dev/null || true)
# #1115: `--dangerously-bypass-hook-trust` only exists on codex-cli >= 0.137.0.
# Capability-probe it so older installs don't fail with "unexpected argument"
# (which, with stderr suppressed, produced a silent empty review). The codex
@@ -281,9 +287,9 @@ fi
**Claude (separate session):**
```bash
if [ -n "$CLAUDE_MODEL" ] && [ "$CLAUDE_MODEL" != "null" ]; then
cat {run_dir}/gsd-review-prompt.md | claude --model "$CLAUDE_MODEL" -p - 2>/dev/null > {run_dir}/gsd-review-claude.md
cat {run_dir}/gsd-review-prompt.md | claude --model "$CLAUDE_MODEL" $CLAUDE_EFFORT_ARGS -p - 2>/dev/null > {run_dir}/gsd-review-claude.md
else
cat {run_dir}/gsd-review-prompt.md | claude -p - 2>/dev/null > {run_dir}/gsd-review-claude.md
cat {run_dir}/gsd-review-prompt.md | claude $CLAUDE_EFFORT_ARGS -p - 2>/dev/null > {run_dir}/gsd-review-claude.md
fi
```
@@ -298,9 +304,9 @@ fi
# stdout redirect would append that noise to a non-empty file — slipping past the
# `[ ! -s … ]` empty-output guard as a silently polluted review.
if [ -n "$CODEX_MODEL" ] && [ "$CODEX_MODEL" != "null" ]; then
cat {run_dir}/gsd-review-prompt.md | codex exec --ephemeral $CODEX_BYPASS_FLAG --model "$CODEX_MODEL" --skip-git-repo-check -o {run_dir}/gsd-review-codex.md - 2>{run_dir}/gsd-review-codex.err >/dev/null
cat {run_dir}/gsd-review-prompt.md | codex exec --ephemeral $CODEX_BYPASS_FLAG --model "$CODEX_MODEL" $CODEX_EFFORT_ARGS --skip-git-repo-check -o {run_dir}/gsd-review-codex.md - 2>{run_dir}/gsd-review-codex.err >/dev/null
else
cat {run_dir}/gsd-review-prompt.md | codex exec --ephemeral $CODEX_BYPASS_FLAG --skip-git-repo-check -o {run_dir}/gsd-review-codex.md - 2>{run_dir}/gsd-review-codex.err >/dev/null
cat {run_dir}/gsd-review-prompt.md | codex exec --ephemeral $CODEX_BYPASS_FLAG $CODEX_EFFORT_ARGS --skip-git-repo-check -o {run_dir}/gsd-review-codex.md - 2>{run_dir}/gsd-review-codex.err >/dev/null
fi
if [ ! -s {run_dir}/gsd-review-codex.md ]; then
echo "Codex review failed or returned empty output. stderr:" > {run_dir}/gsd-review-codex.md
@@ -345,7 +351,7 @@ if [ -n "$OPENCODE_MODEL" ] && [ "$OPENCODE_MODEL" != "null" ]; then
else
set --
fi
cat {run_dir}/gsd-review-prompt.md | opencode run "$@" --format json - 2>{run_dir}/gsd-review-opencode.err > {run_dir}/gsd-review-opencode.json
cat {run_dir}/gsd-review-prompt.md | opencode run "$@" $OPENCODE_EFFORT_ARGS --format json - 2>{run_dir}/gsd-review-opencode.err > {run_dir}/gsd-review-opencode.json
# Reconstruct the review from the assistant text parts into a variable and test
# its CONTENT, not the file size: an empty extraction still prints a trailing
# newline that would fool a `[ -s file ]` check into skipping the stub.