From chama
Run Critical Gate analysis on working tree or specific commit
How this skill is triggered — by the user, by Claude, or both
Slash command
/chama:gate-checkThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Run the Critical Gate engine to analyze diffs for destructive/dangerous operations. This is an **informational** mode — it reports findings but does not block operations.
Run the Critical Gate engine to analyze diffs for destructive/dangerous operations. This is an informational mode — it reports findings but does not block operations.
Read project.language from .chama.yml. Respond in the configured language. Default: pt-BR.
git diff)--commit <commitID>: analyzes the diff of a specific commit vs its predecessorif [ -d "chama/scripts" ]; then
GATE_SCRIPT="chama/scripts/run-critical-gate.sh"
elif [ -d "${HOME}/.claude/plugins/chama/scripts" ]; then
GATE_SCRIPT="${HOME}/.claude/plugins/chama/scripts/run-critical-gate.sh"
else
GATE_SCRIPT="scripts/run-critical-gate.sh"
fi
If the user provided a --commit <commitID> argument:
bash "$GATE_SCRIPT" --mode standalone --commit <commitID>
GATE_EXIT=$?
If no argument was provided (working tree diff):
bash "$GATE_SCRIPT" --mode standalone
GATE_EXIT=$?
Present the gate output to the user with context:
0 (clean): report that no critical operations were detected.1 (CRITICAL/HIGH findings): show all findings with severity, rule ID, file, line, and message. Explain what each finding means and suggest how to address it.2 (warnings only): show warnings and explain their nature. Note that these are non-blocking.3 (error): report the error and suggest troubleshooting steps (e.g., check .chama.yml configuration, verify yq is installed).npx claudepluginhub rafaelportugal/chama --plugin chamaRuns a gate-oriented safety audit on code changes before they land, using a two-pass checklist to triage critical (blocking) and informational issues. Triggers on phrases like "is this safe to land" or "/pre-landing-review".
Reviews git changes with evidence-backed findings and risk-aware verdicts. Supports commit, range, file-scoped analysis, impact assessment, breaking-change detection.
Performs code reviews using Codex MCP with severity-grouped findings and a merge gate. Supports fast (diff-only), full (diff + lint/build), and branch review variants.