feat(agents): template playbooks adopt real request_input defaults - #5211
Conversation
The #5190 follow-up: the elicitation form now has a real default field (shipped in #5177), so the playbook layer retires its two workarounds — "put the recommended value in the field description" and "the FIRST enum option signals the default". - All 28 template playbooks migrated: proposed values become field defaults (one-click accept); researchable enums set "figure it out" AS the default and lean on the built-in Other… escape hatch; fields that already implied multiplicity use the multi-select array shape. Every body stays within its 1-2 KB budget. - write-template-playbooks skill: platform fact #1 flipped (defaults exist — use them; type-matched, empty stripped, dates ignore them), skeleton + checklist + worked example updated. - playbook-spec: "The request_input reality (no defaults)" replaced with the shipped dialect (defaults, multi-select, oneOf choice cards); open-questions #2 closed as shipped-and-adopted. - build-an-agent skill loop step 1 now tells the builder to propose defaults and use the richer shapes at the decisive prompt position.
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (2)
✅ Files skipped from review due to trivial changes (1)
🚧 Files skipped from review as they are similar to previous changes (1)
📝 WalkthroughSummary by CodeRabbit
WalkthroughThe PR updates request_input authoring guidance, playbook specifications, built-in agent-building instructions, and embedded agent templates to use explicit defaults and richer enum, multi-select, and oneOf option shapes. ChangesRequest input contract and authoring guidance
Estimated code review effort: 3 (Moderate) | ~20 minutes Possibly related issues
Possibly related PRs
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
@coderabbitai review |
dosubot
Bot
added
enhancement
✅ Action performedReview finished.
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
.agents/skills/write-template-playbooks/references/prompting-checklist.md (1)
64-65: 🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick winStale field-type guidance contradicts the new multi-select array support.
Lines 64–65 state "No arrays, no nested objects, no multi-select," but
SKILL.mdlines 94–95 explicitly define multi-pick questions as{type: "array", items: {type: "string", enum: [...]}}, and multiple templates inknowledge.py(lines 32–33, 157–158, 217–219, 269–270) now use array-based multi-select defaults. This checklist is the reference playbook authors consult for form elicitation rules — the contradiction will cause authors to avoid the multi-select pattern this PR introduces.🔄 Proposed fix to align with the new multi-select guidance
-Field types are `string`, `number`, `integer`, `boolean` only. No arrays, no nested objects, -no multi-select. Formats: `date`, `date-time`, `email`, `uri`, `multiline`. +Field types are `string`, `number`, `integer`, `boolean`, and `array` (multi-pick only; +use `{type: "array", items: {type: "string", enum: [...]}}`). No nested objects. Formats: +`date`, `date-time`, `email`, `uri`, `multiline`..agents/skills/write-template-playbooks/references/worked-example.md (1)
83-84: 🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win"What this example demonstrates" summary contradicts the updated example body.
The example body (lines 20–23) now instructs setting field defaults and using enum
defaultvalues, but the summary at lines 83–84 still says "No prefilled defaults" and "The recommended enum choice is listed first." This is the canonical example authors copy from — the stale summary directly contradicts the PR's core migration objective and will mislead authors into the old pattern.🔄 Proposed fix to align the summary with the updated example
-- **No prefilled defaults.** The repo guess lives in the field description, not a default field. - The recommended enum choice is listed first. +- **Defaults via the `default` field.** The repo guess is set as the field `default` when a + prior read surfaces one; the recommended enum choice is set as the `default`, not merely + listed first.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro Plus
Run ID: 145b60f0-7d6e-458e-a492-154e4c7fe884
📒 Files selected for processing (12)
.agents/skills/write-template-playbooks/SKILL.md.agents/skills/write-template-playbooks/references/prompting-checklist.md.agents/skills/write-template-playbooks/references/worked-example.mddocs/design/agent-workflows/projects/agent-templates/open-questions.mddocs/design/agent-workflows/projects/agent-templates/playbook-spec.mdsdks/python/agenta/sdk/agents/adapters/agent_templates/engineering.pysdks/python/agenta/sdk/agents/adapters/agent_templates/knowledge.pysdks/python/agenta/sdk/agents/adapters/agent_templates/monitoring.pysdks/python/agenta/sdk/agents/adapters/agent_templates/ops.pysdks/python/agenta/sdk/agents/adapters/agent_templates/sales.pysdks/python/agenta/sdk/agents/adapters/agent_templates/support.pysdks/python/agenta/sdk/agents/adapters/agenta_builtins.py
CodeRabbit caught two passages my phrase-sweep missed (the checklist's field-types line still banning arrays/multi-select, and the worked example's summary still claiming no prefilled defaults); a re-read caught a third in the worked example's intro. All three now state the shipped dialect.
|
@coderabbitai review |
✅ Action performedReview finished.
|
2 participants
Context
The stated follow-up of #5190 (shipped in #5177): the elicitation form now supports a real
defaultper field, multi-select arrays, andoneOfchoice cards — but the template playbooks still taught the pre-#5190 workarounds ("Recommended: " in descriptions, first-enum-option-as-default). Until this lands, builder agents never author the shapes the form can render.What changes
defaults (one-click accept); researchable enums set "Figure it out" as the default and rely on the built-in Other… escape hatch; fields already implying multiplicity use{type: "array", items: {type: "string", enum: [...]}}. Bodies stay within the 1–2 KB budget (largest drift +3%).oneOfwhen the question calls for them.Tests
test_static_catalog: 41 pass (skill-body pins intact)test_agenta_builtins_reference_files: 22 pass (no test pinned the old workaround text)What to QA
Build an agent from any template card: the clarify form should arrive prefilled (accept-in-one-click when the proposals are right), researchable questions should default to "figure it out", and multi-pick questions (e.g. newsletter sources, digest contents) should render as chip pickers.