feat: add agentic team spawning skill router#2060
Open
webguru-hypha wants to merge 2 commits intomainfrom
Open
feat: add agentic team spawning skill router#2060webguru-hypha wants to merge 2 commits intomainfrom
webguru-hypha wants to merge 2 commits intomainfrom
Conversation
Co-authored-by: webguru-hypha <webguru-hypha@users.noreply.github.com>
Contributor
WalkthroughAdded an "agentic team spawning" skill and reference docs that define intent-to-role mapping, team-composition recipes, and a spawning/orchestration workflow with guardrails. Registered the new skill in the skills registry ( Changes
Sequence Diagram(s)sequenceDiagram
participant User
participant Orchestrator
participant RoleMapper
participant Spawner
participant SubAgent
participant Synthesizer
participant MetaArb
User->>Orchestrator: request (task, optional required_roles/exclude_roles/max_agents/focus)
Orchestrator->>RoleMapper: extract intent signals & map to minimal roles
RoleMapper-->>Orchestrator: chosen team + spawn plan
Orchestrator->>Spawner: spawn wave (<=4 concurrent or max_agents)
Spawner->>SubAgent: run sub-agent(s) (assumptions, changes, tests, risks, handoff)
SubAgent-->>Spawner: structured outputs
Spawner-->>Synthesizer: gather wave results
Synthesizer->>Synthesizer: order deps, resolve contradictions
alt unresolved conflicts
Synthesizer->>MetaArb: escalate for arbitration
MetaArb-->>Synthesizer: arbitration result
end
Synthesizer-->>Orchestrator: synthesized plan + validation summary
Orchestrator-->>User: orchestrator response (understanding, team, spawn plan, execution, validation)
Estimated code review effort🎯 1 (Trivial) | ⏱️ ~3 minutes Possibly related PRs
Suggested reviewers
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ 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[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 3
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.agents/skills/agentic-team-spawning/references/role-intent-map.md:
- Around line 26-28: Rewrite rules 3–5 to avoid repeating the "Add ..." starter:
change "Add QA..." to "Include QA when the task requests test automation,
acceptance evidence, or has broad blast radius," change "Add Security..." to
"Include Security for auth, data exposure, untrusted input, secrets, or
compliance-sensitive surfaces," and change "Add Requirements first..." to
"Prioritize Requirements when acceptance criteria are missing and coding would
be speculative," so the three list items read with varied, concise verbs for
easier scanning.
In @.agents/skills/agentic-team-spawning/SKILL.md:
- Around line 63-64: Clarify that the hard concurrency cap of 4 takes precedence
over the user-configurable max_agents: update the SKILL.md text near the "Spawn
at most 4 sub-agents concurrently" line and the `max_agents` description to
state that any `max_agents` value > 4 will be clamped to 4 and the remaining
agents will be executed in additional waves (or, alternatively, mention that
values > 4 are rejected if you prefer validation); reference `max_agents` and
the "Spawn at most 4 sub-agents concurrently" sentence so readers know the cap
is enforced and how overflow agents are handled.
- Line 3: Update the SKILL.md frontmatter `description` to include when this
skill should be invoked (trigger conditions) in addition to what it does: keep
the current summary ("Route a user request to the minimal expert team using
roles in .agents/roles, then spawn only the required sub-agents for execution")
and append a short trigger clause like "Use when a user's request requires
multiple specialized roles or when you need to compose a minimal, focused team
for execution" and optionally include examples (e.g., "invoked for multi-domain
tasks, complex workflows, or when minimizing active agents is required"); edit
the `description` field in .agents/skills/agentic-team-spawning/SKILL.md
accordingly so discoverability and auto-selection logic can match on both
purpose and invocation scenarios.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: 293d8804-e92e-4e42-9bff-8cde420c12bb
📒 Files selected for processing (4)
.agents/skills/agentic-team-spawning/SKILL.md.agents/skills/agentic-team-spawning/references/role-intent-map.md.agents/skills/agentic-team-spawning/references/team-composition-recipes.mdskills-lock.json
Co-authored-by: webguru-hypha <webguru-hypha@users.noreply.github.com>
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.agents/skills/agentic-team-spawning/SKILL.md:
- Around line 120-122: The three guardrail bullets all start with "Never", which
trips static analysis for repetition; edit the lines in SKILL.md that read
"Never spawn 'all experts' by default.", "Never assign the same responsibility
to multiple agents.", and "Never proceed with high-ambiguity implementation
without requirement clarification or triage-first mode." to vary the sentence
starters (e.g., change one to "Do not spawn...", another to "Avoid assigning the
same responsibility...", or "Require clarification or triage-first mode before
proceeding with high-ambiguity implementations") so meaning stays identical but
wording is diversified for easier scanning.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: 42541674-2f68-4d83-b36c-9541a3e1ed8d
📒 Files selected for processing (2)
.agents/skills/agentic-team-spawning/SKILL.md.agents/skills/agentic-team-spawning/references/role-intent-map.md
Comment on lines
+120
to
+122
| - Never spawn "all experts" by default. | ||
| - Never assign the same responsibility to multiple agents. | ||
| - Never proceed with high-ambiguity implementation without requirement clarification or triage-first mode. |
Contributor
There was a problem hiding this comment.
🧹 Nitpick | 🔵 Trivial
Optional: vary the guardrail sentence starters.
Three successive guardrails begin with "Never", which is flagged by static analysis. While the meaning is clear, consider rewording one for variety and easier scanning.
♻️ Optional rephrasing
- Never spawn "all experts" by default.
- Never assign the same responsibility to multiple agents.
-- Never proceed with high-ambiguity implementation without requirement clarification or triage-first mode.
+- Require requirement clarification or triage-first mode before proceeding with high-ambiguity implementation.
- Prefer direct execution without sub-agents if the task is simple and low risk.🧰 Tools
🪛 LanguageTool
[style] ~122-~122: Three successive sentences begin with the same word. Consider rewording the sentence or use a thesaurus to find a synonym.
Context: ...me responsibility to multiple agents. - Never proceed with high-ambiguity implementat...
(ENGLISH_WORD_REPEAT_BEGINNING_RULE)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In @.agents/skills/agentic-team-spawning/SKILL.md around lines 120 - 122, The
three guardrail bullets all start with "Never", which trips static analysis for
repetition; edit the lines in SKILL.md that read "Never spawn 'all experts' by
default.", "Never assign the same responsibility to multiple agents.", and
"Never proceed with high-ambiguity implementation without requirement
clarification or triage-first mode." to vary the sentence starters (e.g., change
one to "Do not spawn...", another to "Avoid assigning the same
responsibility...", or "Require clarification or triage-first mode before
proceeding with high-ambiguity implementations") so meaning stays identical but
wording is diversified for easier scanning.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
.agents/skills/agentic-team-spawning/SKILL.mdto orchestrate minimal expert-team spawning based on user prompt intent.agents/skills/agentic-team-spawning/references/role-intent-map.md.agents/skills/agentic-team-spawning/references/team-composition-recipes.mdskills-lock.jsonValidation
pnpm run format:fixwc -l .agents/skills/agentic-team-spawning/SKILL.md(confirmed < 200 lines)references/*linked fromSKILL.mdgit status --short(clean)CodeRabbit Autofix Notes
Resolved suggestions implemented:
descriptioninSKILL.mdto include trigger conditions.max_agents(min(max_agents, 4)+ wave scheduling).role-intent-map.mdto avoid repetitive sentence starters while preserving meaning.Summary by CodeRabbit