| name | loop-factory |
| description | Use when orchestrating Loop Factory specs from factory/specs/inbox or factory/specs/active, dispatching work to Codex or Claude agents, generating implementation prompts, or deciding next loop state. |
Loop Factory
Use this skill to run spec-driven agent loops.
Workflow
- For new specs, use
grill-me pattern before dispatch: ask one question at a time, provide recommended answer, and record final decisions in # Grill Gate.
- Inspect queue:
python3 bin/loop-factory scan
- Choose one inbox spec unless user asks for batch work.
- Generate prompt:
python3 bin/loop-factory dispatch --agent codex --limit 1
- If implementation starts, stage spec:
python3 bin/loop-factory dispatch --agent codex --limit 1 --stage
- Implement only acceptance criteria.
- Run verification from spec frontmatter.
- Leave spec active for review.
Agent Use
For parallel work, explicitly ask Codex to spawn subagents. Use:
spec-implementer for scoped implementation
spec-reviewer for independent review
spec-backpropagator for spec/doc updates from diffs
Do not fan out same-file edits. Do not automate product choices. If spec is ambiguous, record assumption or open question.
Built-In Loops
Use repo-native loops when user asks for repeated work:
- list:
python3 bin/loop-factory loops list
- inspect:
python3 bin/loop-factory loops show <loop-id>
- render:
python3 bin/loop-factory loops prompt <loop-id> --agent codex
Prefer spec-grill-gate, active-spec-verify, independent-review-gate, and backprop-spec-sync for normal spec flow.
Outputs
Report:
- spec id
- files changed
- verification commands and results
- remaining risks
- whether review/archive is ready