Menú
Diagnostic and recovery playbook for resolving local git states, locked configuration files, broken workspace setups, and process deadlocks.
Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.
Basado en la clasificación ocupacional SOC
Playbook for capturing design alignment from /grill-me, generating issue specifications, and managing task boards.
Playbook instructing agents how to dynamically formulate, design, bootstrap, and register new workspace skills when facing skill gaps.
Playbook for translating natural language requests about tasks, profiles, locking, and validation into CLI helper executions.
Playbook for writing safe database migrations, managing schema evolutions, executing reversible rollbacks, and avoiding table lock contention in enterprise environments.
Guidelines for CPU profiling, identifying database query bottlenecks (N+1 queries), diagnosing memory leaks, and optimizing resource execution speeds.
Guidelines for containerization (Dockerfile best practices), release versioning, blue-green deployment, feature flag rollouts, and post-deployment smoke verification.
| name | self-healing |
| description | Diagnostic and recovery playbook for resolving local git states, locked configuration files, broken workspace setups, and process deadlocks. |
This playbook defines standard operational procedures for diagnosing, recovering, and healing a corrupted developer or agent workspace environment without manual intervention.
When an autonomous agent runs into Git anomalies (detached HEAD, dirty tree, merge conflicts, or lock file issues), it must apply these recovery steps.
git status and search for files with unmerged paths (marked with UU).git merge --abort
<<<<<<<, =======, >>>>>>>), resolve the logic, and stage the file:
git add <resolved_file>
If a git action returns Fatal: Unable to create '.git/index.lock': File exists, the agent should:
rm -f .git/index.lock
Stale locks on modules or databases can freeze agent pipelines.
If a file remains locked by a dead or non-existent branch:
.agents/locks.json to identify the lock holder.git show-ref refs/heads/<holder_branch>
python3 .agents/scripts/cli/helper.py lock <module_name> --release
If an agentic execution fails because of missing dependencies, outdated libraries, or environment compiler errors:
npm install --no-audit or npm ci.pip install -r requirements.txt.find . -name "*.pyc" -delete && find . -name "__pycache__" -deleterm -rf .next build dist./helper.sh bootstrap to restore default stack configuration guidelines.