// Diagnose the cc-suite setup in the current project. Runs the full health check, explains every issue, and fixes what can be fixed automatically. Skill counterpart to /cc-suite:diagnose.
| name | diagnose |
| description | Diagnose the cc-suite setup in the current project. Runs the full health check, explains every issue, and fixes what can be fixed automatically. Skill counterpart to /cc-suite:diagnose. |
| version | 0.2.6 |
Run a full cc-suite health check and offer to auto-fix every issue found. Skill counterpart to /cc-suite:diagnose.
/cc-suite:audit, /cc-suite:audit-fix, or any /cc-suite:claude-* skill behaves unexpectedlyRun the status script:
bash "${CLAUDE_PLUGIN_ROOT}/scripts/status.sh"
Stale nested symlinks (macOS ln -sf residue):
find -L .claude/skills/ .agents/skills/ -maxdepth 3 -name "cc-suite" -type d 2>/dev/null \
| grep -vE "^\.claude/skills/cc-suite$|^\.agents/skills/cc-suite$"
Any output is a stale nested symlink that duplicates skills in Codex.
Codex CLI binary:
which codex 2>/dev/null || echo "not-found"
Cache freshness — extract the version from the active symlink target and compare to ${CLAUDE_PLUGIN_ROOT}/../../.claude-plugin/plugin.json (the installed plugin version). If they differ, the skills symlink points to an old cache.
readlink .claude/skills/cc-suite 2>/dev/null | grep -oE '[0-9]+\.[0-9]+\.[0-9]+' || echo "missing"
Broken symlinks:
[ -L .claude/skills/cc-suite ] && [ ! -d .claude/skills/cc-suite ] && echo "broken:claude-skills" || true
[ -L .agents/skills ] && [ ! -d .agents/skills ] && echo "broken:agents-skills" || true
Build an issues list from all · (missing) and ! (warn) lines in the status output, plus any deep check failures.
Exclude .codex/hooks.json from issues if the project has no hooks section in .claude/settings.json — that missing entry is expected.
Display the diagnosis:
cc-suite diagnose — {cwd}
Healthy: N items ✓
Issues found: N
# Item Status Diagnosis
1 .agents/skills missing Codex cannot see any skills
2 .mcp.json → codex-cli missing Claude cannot invoke Codex as MCP tool
3 plugin_hooks not set Plugin-bundled hooks are inert in Codex
...
If no issues: report healthy and stop.
Ask:
Fix all auto-fixable issues now? (yes / show commands only / cancel)
For each fixable issue, run the corresponding script:
| Issue | Fix |
|---|---|
.agents/skills missing or wrong | bash "${CLAUDE_PLUGIN_ROOT}/scripts/bridge_skills.sh" |
.claude/skills/cc-suite missing or wrong | bash "${CLAUDE_PLUGIN_ROOT}/scripts/bridge_skills.sh" |
.codex/hooks.json missing | python3 "${CLAUDE_PLUGIN_ROOT}/scripts/bridge_hooks.py" |
.mcp.json → codex-cli missing | bash "${CLAUDE_PLUGIN_ROOT}/scripts/mcp_codex.sh" |
.codex/config.toml → claude-code missing | bash "${CLAUDE_PLUGIN_ROOT}/scripts/mcp_claude.sh" |
| MCP parity gaps | bash "${CLAUDE_PLUGIN_ROOT}/scripts/bridge_mcp.sh" |
stale nested symlink at {path} | rm "{path}" && bash "${CLAUDE_PLUGIN_ROOT}/scripts/bridge_skills.sh" |
| cache stale | bash "${CLAUDE_PLUGIN_ROOT}/scripts/bridge_skills.sh" (repoints to current cache) |
plugin_hooks not set | Write plugin_hooks = true under [features] in ~/.codex/config.toml |
Items that require manual action (flag, do not attempt to fix):
project trust not trusted — run codex in this directory and accept the trust promptAGENTS.md missing — run /cc-suite:init in a Claude Code session (requires Claude)After all auto-fixes, run bash "${CLAUDE_PLUGIN_ROOT}/scripts/status.sh" again.
Report: N issues fixed, N remaining (with manual steps for those that remain).
If issues persist after auto-fix, close with: "Issues remain. Next step: run /cc-suite:repair for a full non-interactive re-run of all setup scripts. If that also fails, run /cc-suite:init for a complete interactive re-initialization."
Full audit→fix→verify cycle: Claude audits, Codex fixes, Claude verifies. Repeats up to 3 rounds until all issues are resolved or the user stops. Claude does all code reading and judgment; Codex does all file editing.
Initialize cc-suite for the current project — sets up the AGENTS.md bridge, registers Codex and Claude MCP servers, and generates a .cc-suite.md config. Skill counterpart to /cc-suite:init.
Non-interactive re-run of all cc-suite bridge and registration scripts. No questions asked — idempotent escalation step after the diagnose skill finds issues it could not fix. Skill counterpart to /cc-suite:repair.
Use when writing, reviewing, or naming any cc-suite artifact — pick the canonical noun or verb from this registry rather than coining a synonym. Loaded by NLPM's scorer and checker when R51 is enabled in .claude/nlpm.local.md.
Use when creating, editing, or reviewing a cc-suite advisor agent (a value-over-rules persona under `.cc-suite/agents/`). Covers system-prompt phrasing, model choice, tool restrictions, working-dir scoping, budget/turn limits, and the file format. Use proactively whenever the user asks to "add an agent", "make an advisor", "write a reviewer", or edits anything under `.cc-suite/agents/`.
Send a bug or failing test to Claude Code for root-cause analysis and fix. Use when you've hit an error you can't trace, a test is failing for unclear reasons, or a behavior is wrong and the cause is deep in the codebase.