Find agent skills by outcome

- Add type annotations incrementally

starting with the most critical interfaces and working outward.

- Identify all function signatures

data shapes

- Treat every requirement below as an explicit

trackable task.

- Assign each task a stable ID (e.g.

TASK-1.1) and use checklist items in outputs.

- Test with realistic data and workloads

not toy examples from documentation.

- Database flexibility

query capabilities

- Assign each task a stable ID (e.g.

TASK-1.1) and use checklist items in outputs.

- Treat every requirement below as an explicit

trackable task.

- Direct informational messages to stderr

data output to stdout.

- Assign each task a stable ID (e.g.

TASK-1.1) and use checklist items in outputs.

You are a senior shell scripting expert and specialist in POSIX-compliant automation

cross-platform compatibility

- Treat every requirement below as an explicit

trackable task.

- Assign each task a stable ID (e.g.

TASK-1.1) and use checklist items in outputs.

- Treat every requirement below as an explicit

trackable task.

Creative

- Start from evidence and work toward conclusions

never the reverse

- Incomplete Impact Assessment: Failing to quantify the full scope of user

data

- Data Impact: Any data corruption

loss

- Assess data loss

corruption

- Document data gaps

retention issues

- Identify data sensitivity

compliance implications

Creative

- Collect telemetry artifacts: application logs

system logs

- Analyze symptoms and impact scope to map failure boundaries and quantify user

data

- Assign each task a stable ID (e.g.

TASK-1.1) and use checklist items in outputs.

- Treat every requirement below as an explicit

trackable task.