Skip to content

great_cto

v2.33.1 Bugfix

This release fixes issues for SREs watching stability and regressions.

Published 5d AI Agents & Assistants
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agentic-coding claude-code-plugin claude-code-skills claude-code-subagents code-review cto
+2 more
multi-agent sdlc

Affected surfaces

auth

Summary

AI summary

Fixed three SessionStart configuration bugs causing command failures and missing agent sync.

Changes in this release

Bugfix Medium

Fixes SessionStart config bugs in plugin.json (bash/sh syntax, model-override path, missing agents).

Fixes SessionStart config bugs in plugin.json (bash/sh syntax, model-override path, missing agents).

Source: llm_adapter@2026-05-29

Confidence: low

 —
Bugfix Medium

Corrects bash/sh syntax error caused by stray `;` in SessionStart command.

Corrects bash/sh syntax error caused by stray `;` in SessionStart command.

Source: granite4.1:30b@2026-05-29-audit

Confidence: low

 —
Bugfix Medium

Fixes broken model-override script path using `${PLUGIN_DIR}`.

Fixes broken model-override script path using `${PLUGIN_DIR}`.

Source: granite4.1:30b@2026-05-29-audit

Confidence: low

 —
Bugfix Medium

Ensures five routed agents are synced and written to `~/.claude/agents/`.

Ensures five routed agents are synced and written to `~/.claude/agents/`.

Source: granite4.1:30b@2026-05-29-audit

Confidence: low

 —
Full changelog

Fixed: 3 SessionStart config bugs in plugin.json

Closes the remaining 3 bug-hunt findings that were tracked in
docs/PENDING-v2.32.1.md (the other 5 — hooks + CLI — already shipped in
v2.33.0 via commit 448ad48). All three live in the SessionStart hook
command in .claude-plugin/plugin.json:

  • #6 — &; broke SessionStart under bash/sh. Six &; sequences
    (background + stray ;) are accepted by zsh but rejected by bash/sh with
    a syntax error that aborted the entire hook (no agent/command sync, no
    PROJECT/brain load). Replaced all six &; → &. bash -n now passes.
  • #4 — broken model-override path. bash "\/scripts/apply-model-override.sh"
    (stray backslash, missing ${PLUGIN_DIR}) never resolved, so
    ~/.great_cto/config: agent-model was silently ignored. Fixed to
    "${PLUGIN_DIR}/scripts/apply-model-override.sh".
  • #2 — 5 routed agents never synced. gdpr-reviewer,
    us-privacy-reviewer, dpdpa-reviewer, digital-health-reviewer, and
    coordinator were absent from the agent copy-list, so they were never
    written to ~/.claude/agents/ and dispatch silently failed. Appended them.

This completes the v2.32.0 bug-hunt (9 findings, all verified and closed).
Version files re-synced together (plugin.json + package.json + jsr.json) —
the original v2.32.0 drift came from a partial bump.

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track great_cto

Get notified when new releases ship.

Sign up free

About great_cto

Engineering-management layer of 34 specialist AI agents covering the full SDLC (architect, PM, senior-dev, reviewer, QA, security, devops, L3-support + 18 archetype-specific reviewers) with auto-detected archetypes and compliance gates (PCI-DSS, HIPAA, FedRAMP, GDPR, EU AI Act). Runs in Claude Code, Cursor, Codex CLI, Aider, and Continue via AGENTS.md + MCP. MIT.

All releases →

Related context

Earlier breaking changes

  • v2.32.0 Removed AgentShield scanner and its CLI commands.

Beta — feedback welcome: [email protected]