Adonis0123/adonis-skills / skills/workflow-gate
workflow-gate
VerifiedClassify which workflow or skill should run when route choice matters: unresolved creative work, named-option decisions, explicit Goal management, destructive actions, architecture work, planning, debugging or ship checks, review handoff, and full-completion pipelines. Skip trivial one-line work and clearly matching safe named-skill requests. Treat first-principles analysis as a method, not a Challenge trigger unless the user is widening the option space.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 13.2 KB |
| agents/openai.yaml | config | 410 B |
| evals/evals.json | config | 41.7 KB |
| evals/trigger-eval.json | config | 3.1 KB |
| references/examples-core.md | markdown | 5.2 KB |
| references/examples-edge.md | markdown | 14.9 KB |
| references/route-adjustments.md | markdown | 3.7 KB |
| references/workflow-systems.md | markdown | 9.5 KB |
references/route-adjustments.md
View on GitHub# Route adjustments
Fallback-only. Do not load this file for Architecture / Review-Handoff / Challenge — those already live in SKILL.md precedence + tiebreakers. Use this only when the Route picked from the cheat card still doesn't fit on closer inspection. Each rule names _why_ it exists so you can generalize when the literal trigger doesn't match.
## Upgrade — smaller Route is unsafe
- _Any → Light:_ the task involves a file write, verification, or repo mutation. _Why:_ writes are decisions, not lookups.
- _Light → Challenge (convergence):_ behavior ambiguous, output undefined, or 2+ implementations with different tradeoffs. _Why:_ committing without alignment guarantees rework.
- _Any → authorization hold:_ `destructive=yes` (drop table, force push, delete prod data, schema break, public API removal) → `Runtime skill: none`, one blocking authorization question, no work started. _Why:_ failure cost dwarfs one question. (This is Rule #1; restated for completeness.)
- _Any → Review-Handoff:_ user asks for cross-agent / packet review or fix-then-re-review. _Why:_ review skills keep eyes fresh.
- _Light/Plan → Architecture:_ user names a path/module and asks to diagnose structure or run a bounded harden loop. _Why:_ architecture jobs are not ordinary plans or typo-scale edits.
- _Direct/Light → Challenge:_ creative HARD-GATE work without a paid design/spec. _Why:_ implementation without a thesis burns rework.
## Downgrade — heavier Route is ceremony
- _Challenge → Direct/Light:_ the request is not creative HARD-GATE work, or the prompt/spec already fixes the relevant design and behavior. _Why:_ re-challenging wastes their signal only after the design decision has been paid.
- _Challenge (convergence) → Light:_ one reasonable implementation; user supplied exact behavior. _Why:_ alignment is implicit.
- _Plan → Light:_ the immediate request is direct implementation, regardless of file count. _Why:_ scope changes risk and verification depth, not user intent; a user-facing plan would be overhead.
- _Architecture + unresolved scope → Architecture + `Runtime skill: none`:_ first try user-supplied attachment/reference/Git locators plus one cheap repo signal; only then ask one minimal scope question and do not load a scanner. _Why:_ scanners without a frozen file set become whole-repo thrash, while making users repeat already-resolvable scope wastes a full round trip.
- _Architecture → Light/Challenge:_ the ask is really a bug symptom or product decision rather than structure work. _Why:_ bugs belong to systematic-debugging; named-option decisions belong to Challenge convergence.
- _architecture-hardening-loop → improve-codebase-architecture:_ user asked only for diagnosis/report. _Why:_ harden loops mutate code; diagnose must not silently upgrade.
- _Review-Handoff + missing Git repository:_ follow SKILL.md Rule #6 and its failure contract.
## Re-gate mid-task
Re-run the cheat card and precedence rules when the active Route's preconditions change:
- A new destructive signal surfaces (e.g. mid-Plan you find the migration drops a column).
- Scope grows beyond the current plan assumptions or shrinks to one obvious file.
- A new blocking bug appears (Plan → Light + `systematic-debugging` until it closes).
- User adds "don't ask" / "discuss first".
- Architecture diagnose finishes and the user then explicitly asks to harden/implement in the same frozen scope → re-gate to Architecture + `architecture-hardening-loop`.
- Challenge `agent-strawman` is confirmed by the user, or explicitly delegated as a reversible product direction → load grilling / grill-with-docs (do not re-emit a different Route unless signals changed).
Re-gate only when a listed precondition changes; do not re-run it as routine ceremony.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deployment | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy. | Stars: 119k | Updated | Claude CodeCodex | Verified | Install |
| childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |