@aryaminus/controlkeel-opencode 0.3.5 → 0.3.6

package/.opencode/skills/benchmark-operator/SKILL.md

@@ -16,6 +16,8 @@ metadata:
   author: controlkeel
   version: "2.0"
   category: benchmark
+  ck_mcp_tools:
+    - ck_observability
 ---
 
 # Benchmark Operator Skill
@@ -29,6 +31,15 @@ Use this skill when the task is benchmark orchestration instead of normal govern
 3. Review catch rate, block rate, expected-rule hit rate, latency, and overhead.
 4. Export the run if you need external analysis.
 
+## Local observability feedback loop
+
+For generated observability coverage, prefer the human-gated loop:
+
+1. Inspect `ck_observability` reports for `saved_evals`, `benchmark_drafts`, `benchmark_scenarios`, and `benchmark_history`.
+2. Use CLI-only commands for mutations: draft approval, materialization, and benchmark execution are not exposed through MCP in this skill.
+3. Run generated observability benchmarks only after a dry-run review and explicit operator approval.
+4. Treat `promotions` as advisory evidence; do not mutate policy, router, prompt, or autofix artifacts from benchmark results alone.
+
 ## Additional resources
 
 - [Benchmark operator playbook](references/benchmark-playbook.md)

package/.opencode/skills/policy-training/SKILL.md

@@ -17,6 +17,7 @@ metadata:
   version: "2.0"
   category: policy
   ck_mcp_tools:
+    - ck_observability
     - ck_outcome_tracker
 ---
 
@@ -32,6 +33,10 @@ Use this skill only when the task is offline policy training or artifact promoti
 4. Summarize held-out metrics against the heuristic baseline before promotion.
 5. Consider real-world success inputs using `ck_outcome_tracker` (leaderboards, recorded session outcomes).
 
+## Observability promotion evidence
+
+Before policy artifact promotion, inspect `ck_observability` report `promotions` and `benchmark_history`. Promotion candidates are advisory and non-mutating; require human approval, successful held-out or generated benchmark evidence, and explicit policy-training intent before changing router, budget-hint, prompt, or policy artifacts.
+
 ## Additional resources
 
 - [Promotion rules](references/promotion-rules.md)

package/.opencode/skills/proof-memory/SKILL.md

@@ -16,6 +16,7 @@ metadata:
   version: "2.0"
   category: proof
   ck_mcp_tools:
+    - ck_observability
     - ck_context
     - ck_memory_search
     - ck_memory_record
@@ -36,6 +37,10 @@ Use this skill when you need the durable system-of-record view instead of only t
 5. Use `ck_memory_record` to preserve new decisions or operator intent that future agents should recover explicitly.
 6. Use `ck_memory_archive` to retire stale or superseded memories so retrieval quality does not decay.
 
+## Observability evidence
+
+When closing or resuming work, include local observability loop evidence when relevant: saved evals, benchmark drafts, materialized scenarios, generated benchmark history, and advisory promotion candidates. Prefer the read-only `ck_observability` MCP surface for summaries and record durable checkpoints with `ck_memory_record`; do not treat advisory promotion candidates as executed changes.
+
 ## Additional resources
 
 - [Proof workflow](references/proof-workflow.md)

package/.opencode/skills/ship-readiness/SKILL.md

@@ -16,6 +16,7 @@ metadata:
   version: "2.0"
   category: release
   ck_mcp_tools:
+    - ck_observability
     - ck_context
     - ck_deployment_advisor
 ---
@@ -32,6 +33,10 @@ Use this skill when the operator asks whether a mission or session is ready for
 4. Summarize approvals, rejections, and any remaining human work.
 5. Provide automatic deployment resources via `ck_deployment_advisor` (Dockerize, CI pipes) for the relevant stack (Phoenix, etc.).
 
+## Observability readiness
+
+Before calling a feature ready, check whether relevant local observability evidence exists. Use `ck_observability` reports for `benchmark_history` and `promotions` to summarize readiness, uncovered scenarios, missed runs, and advisory promotion candidates. A ready advisory candidate is not a policy/router/prompt promotion; it still requires explicit human review.
+
 ## Additional resources
 
 - [Release checklist](references/release-checklist.md)

package/package.json

@@ -35,5 +35,5 @@
     "url": "git+https://github.com/aryaminus/controlkeel.git"
   },
   "type": "module",
-  "version": "0.3.5"
+  "version": "0.3.6"
 }