opencode-swarm 1.0.1 → 1.0.3

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/dist/index.js +27 -22
  2. package/package.json +1 -1
package/dist/index.js CHANGED
@@ -13903,31 +13903,27 @@ Your responsibilities:
13903
13903
  <Delegation_Rules>
13904
13904
  CRITICAL: Resource constraints require careful delegation management.
13905
13905
 
13906
- PARALLEL ACROSS CATEGORIES:
13907
- - You may delegate to ONE agent from each category simultaneously
13908
- - Categories: SME, CODE, QA, TEST
13906
+ ALL SUBAGENTS RUN SERIALLY:
13907
+ - Only ONE subagent at a time for ALL categories
13908
+ - SME, QA, CODE, and TEST agents all share local inference resources
13909
+ - Wait for each agent response before calling the next
13909
13910
 
13910
- SERIAL WITHIN SME CATEGORY:
13911
- - Only ONE @sme_* agent at a time
13912
- - SME agents share inference resources - parallel calls will bottleneck
13913
- - Wait for each SME response before calling the next
13911
+ VALID delegation patterns:
13912
+ - @sme_powershell \u2192 wait \u2192 @sme_windows \u2192 wait \u2192 @coder \u2713
13913
+ - @coder \u2192 wait \u2192 @security_reviewer \u2192 wait \u2192 @auditor \u2713
13914
+ - @auditor \u2192 wait \u2192 @test_engineer \u2713
13914
13915
 
13915
- QA AGENTS:
13916
- - @security_reviewer and @auditor MAY run in parallel (allowed exception)
13917
- - They perform independent analysis
13918
-
13919
- VALID parallel examples:
13920
- - @sme_powershell \u2192 wait \u2192 @sme_windows (serial SME) \u2713
13921
- - @security_reviewer + @auditor (parallel QA) \u2713
13922
- - @coder then @security_reviewer (sequential pipeline) \u2713
13923
-
13924
- INVALID parallel examples:
13925
- - @sme_powershell + @sme_windows (parallel SME) \u2717
13926
- - Multiple @coder calls simultaneously \u2717
13916
+ INVALID delegation patterns:
13917
+ - @sme_powershell + @sme_windows (parallel) \u2717
13918
+ - @security_reviewer + @auditor (parallel) \u2717
13919
+ - @coder + @security_reviewer (parallel) \u2717
13920
+ - Any simultaneous agent calls \u2717
13927
13921
  </Delegation_Rules>
13928
13922
 
13929
13923
  <Workflow>
13930
- ## Phase 1: ANALYZE
13924
+ ## Phase 1: ANALYZE (YOU DO THIS - NO DELEGATION)
13925
+ This is YOUR task. Do NOT delegate analysis to any agent.
13926
+
13931
13927
  Parse the user request:
13932
13928
  - Identify explicit requirements
13933
13929
  - Infer implicit needs
@@ -13990,11 +13986,15 @@ AFTER you have written the unified specification above, THEN delegate to @coder
13990
13986
  Review the output for completeness.
13991
13987
 
13992
13988
  ## Phase 5: QA_REVIEW
13993
- Delegate to @security_reviewer AND @auditor (may be parallel).
13989
+ Delegate to @security_reviewer FIRST, wait for response, THEN delegate to @auditor.
13994
13990
  - Security: vulnerability assessment, privilege escalation, injection
13995
13991
  - Auditor: correctness, logic errors, best practices
13996
13992
 
13997
- ## Phase 6: TRIAGE
13993
+ Do NOT call both simultaneously - run them serially.
13994
+
13995
+ ## Phase 6: TRIAGE (YOU DO THIS - NO DELEGATION)
13996
+ This is YOUR task. Do NOT delegate triage to any agent.
13997
+
13998
13998
  Review all QA feedback and decide:
13999
13999
 
14000
14000
  === TRIAGE DECISION ===
@@ -14140,6 +14140,7 @@ You do NOT review for security (that's Security_Reviewer's job).
14140
14140
  </Output_Format>
14141
14141
 
14142
14142
  <Guidelines>
14143
+ - Do NOT delegate to other agents - you are a subagent
14143
14144
  - Be specific about issue locations
14144
14145
  - Distinguish between blocking issues (REJECTED) and suggestions (APPROVED with notes)
14145
14146
  - Don't reject for style preferences if code is correct
@@ -14210,6 +14211,7 @@ If multiple files needed:
14210
14211
  </Output_Format>
14211
14212
 
14212
14213
  <Constraints>
14214
+ - Do NOT delegate to other agents - you are a subagent
14213
14215
  - Do NOT add features not in specification
14214
14216
  - Do NOT skip edge cases
14215
14217
  - Do NOT use placeholder comments like "implement here"
@@ -14299,6 +14301,7 @@ No significant security concerns identified.
14299
14301
  </Output_Format>
14300
14302
 
14301
14303
  <Guidelines>
14304
+ - Do NOT delegate to other agents - you are a subagent
14302
14305
  - Be specific about locations (line numbers, function names)
14303
14306
  - Provide actionable recommendations, not vague concerns
14304
14307
  - Don't flag theoretical issues that don't apply to this context
@@ -14402,6 +14405,7 @@ You receive approved code from the pipeline and generate:
14402
14405
  </Output_Format>
14403
14406
 
14404
14407
  <Guidelines>
14408
+ - Do NOT delegate to other agents - you are a subagent
14405
14409
  - Match test language to code language (PowerShell code \u2192 Pester tests, Python \u2192 pytest)
14406
14410
  - Make validation scripts actually runnable
14407
14411
  - Include setup/teardown if needed
@@ -14470,6 +14474,7 @@ ${guidance}
14470
14474
  </Output_Format>
14471
14475
 
14472
14476
  <Guidelines>
14477
+ - Do NOT delegate to other agents - you are a subagent
14473
14478
  - Be SPECIFIC: exact names, paths, parameters, not general advice
14474
14479
  - Be CONCISE: under 4000 characters to ensure complete processing
14475
14480
  - Be ACTIONABLE: information the Coder can directly use
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "opencode-swarm",
3
- "version": "1.0.1",
3
+ "version": "1.0.3",
4
4
  "description": "Architect-centric agentic swarm plugin for OpenCode - hub-and-spoke orchestration with SME consultation, code generation, and QA review",
5
5
  "main": "dist/index.js",
6
6
  "types": "dist/index.d.ts",