@lythos/skill-arena 0.15.3 → 0.15.5

package/README.md

@@ -16,20 +16,20 @@
 ```bash
 bun add -d @lythos/skill-arena
 # or use directly
-bunx @lythos/skill-arena@0.15.3 <command>
+bunx @lythos/skill-arena@0.15.5 <command>
 ```
 
 ## Quick Start
 
 ```bash
 # single — test one deck (most common)
-bunx @lythos/skill-arena@0.15.3 single \
+bunx @lythos/skill-arena@0.15.5 single \
   --deck ./examples/decks/scout.toml \
   --brief "Generate auth flow diagram" \
   --out ./output
 
 # single with explicit player
-bunx @lythos/skill-arena@0.15.3 single \
+bunx @lythos/skill-arena@0.15.5 single \
   --deck ./examples/decks/scout.toml \
   --brief "Generate auth flow diagram" \
   --player kimi \
@@ -37,10 +37,10 @@ bunx @lythos/skill-arena@0.15.3 single \
 
 # cross-deck vs — compare two decks (agent-orchestrated)
 # Create arena.toml declaring sides with different decks, then:
-bunx @lythos/skill-arena@0.15.3 vs --config ./arena.toml
+bunx @lythos/skill-arena@0.15.5 vs --config ./arena.toml
 
 # cross-player vs — compare kimi vs codex (CLI only)
-bunx @lythos/skill-arena@0.15.3 vs --config ./arena.toml --player kimi
+bunx @lythos/skill-arena@0.15.5 vs --config ./arena.toml --player kimi
 ```
 
 **What happens**: Agent creates isolated `/tmp` workdir per side, `deck link` skills, spawns parallel subagents, collects artifacts, judge scores outputs. Parent deck restored after.
@@ -50,14 +50,14 @@ bunx @lythos/skill-arena@0.15.3 vs --config ./arena.toml --player kimi
 ### `single` — one deck, one task
 
 ```bash
-bunx @lythos/skill-arena@0.15.3 single \
+bunx @lythos/skill-arena@0.15.5 single \
   --deck ./deck.toml \
   --brief "Produce a .docx report with radar chart" \
   --timeout 600000 \
   --out ./output
 
 # with explicit player
-bunx @lythos/skill-arena@0.15.3 single \
+bunx @lythos/skill-arena@0.15.5 single \
   --deck ./deck.toml \
   --brief "Produce a .docx report with radar chart" \
   --player kimi \
@@ -67,14 +67,14 @@ bunx @lythos/skill-arena@0.15.3 single \
 ### `vs` — multi-deck comparison
 
 ```bash
-bunx @lythos/skill-arena@0.15.3 vs --config ./arena.toml
-bunx @lythos/skill-arena@0.15.3 vs --config ./arena.toml --dry-run
+bunx @lythos/skill-arena@0.15.5 vs --config ./arena.toml
+bunx @lythos/skill-arena@0.15.5 vs --config ./arena.toml --dry-run
 ```
 
 ### `prepare-workdir` — isolate + link skills (agent-orchestrated)
 
 ```bash
-bunx @lythos/skill-arena@0.15.3 prepare-workdir \
+bunx @lythos/skill-arena@0.15.5 prepare-workdir \
   --deck ./skill-deck.toml \
   --out /tmp/arena-side-a \
   --brief "task description"
@@ -85,7 +85,7 @@ Creates `/tmp`-isolated workdir with deck copied, AGENTS.md written, and `deck l
 ### `archive` — collect agent outputs (agent-orchestrated)
 
 ```bash
-bunx @lythos/skill-arena@0.15.3 archive \
+bunx @lythos/skill-arena@0.15.5 archive \
   --from /tmp/arena-side-a \
   --to ./playground/output \
   --sides side-a
@@ -96,7 +96,7 @@ Copies agent artifacts from workdir(s) to output, skipping internal files (`.cla
 ### `viz` — render results (WIP — HTML report generation pending)
 
 ```bash
-bunx @lythos/skill-arena@0.15.3 viz runs/arena-<id>/
+bunx @lythos/skill-arena@0.15.5 viz runs/arena-<id>/
 ```
 
 ## Parameters

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lythos/skill-arena",
-  "version": "0.15.3",
+  "version": "0.15.5",
   "description": "Skill Arena — benchmark skill effectiveness with controlled-variable comparison",
   "keywords": [
     "ai-agent",
@@ -42,15 +42,15 @@
     "bun": ">=1.0.0"
   },
   "dependencies": {
-    "@lythos/cold-pool": "^0.15.3",
-    "@lythos/infra": "^0.15.3",
-    "@lythos/test-utils": "^0.15.3",
+    "@lythos/cold-pool": "^0.15.5",
+    "@lythos/infra": "^0.15.5",
+    "@lythos/test-utils": "^0.15.5",
     "zod": "^3.24.0",
     "zod-to-json-schema": "^3.25.2"
   },
   "optionalDependencies": {
-    "@lythos/agent-adapter-claude-sdk": "^0.15.3",
-    "@lythos/agent-adapter-deepseek-serve": "^0.15.3",
-    "@lythos/agent-adapter-codex": "^0.15.3"
+    "@lythos/agent-adapter-claude-sdk": "^0.15.5",
+    "@lythos/agent-adapter-deepseek-serve": "^0.15.5",
+    "@lythos/agent-adapter-codex": "^0.15.5"
   }
 }

package/src/preflight.ts

@@ -276,14 +276,14 @@ export function buildPreparePlan(params: {
     '',
     '## Setup Order (why this sequence)',
     '1. `skill-deck.toml` copied here → declares which skills you can use',
-    '2. `deck link` runs → cold pool skills become visible in `.claude/skills/`',
+    '2. `deck link` runs → cold pool skills become visible in the working set',
     '3. Skill existence checked → warns if any declared skill is missing from cold pool',
     '4. `AGENTS.md` written last → confirms setup succeeded before agent starts',
     'If setup fails mid-sequence, the workdir is incomplete and nothing runs.',
     '',
     '## How This Works',
     '- Write ALL output files to this directory (CWD).',
-    '- Use available skills — check `ls .claude/skills/`.',
+    '- Use available skills — check the working set directory (e.g. `ls .claude/skills/`).',
     '',
     '## Output Contract',
     '- MANDATORY: `decision-log.jsonl` — one JSON line per decision:',

package/src/runner.ts

@@ -84,7 +84,7 @@ export function buildArenaPrompt(opts: {
     'ROBUSTNESS — If any command or script fails, read the error output, fix the issue, and retry.',
     'Do not stop on the first error. Ensure all required output files exist before finishing.',
     '',
-    'TOOLS — Use the skills already linked in .claude/skills/ (check with `ls .claude/skills/`).',
+    'TOOLS — Use the skills already linked in your working set (check with `ls .claude/skills/` or your configured path).',
     'They are available and tested. Only write alternative scripts if the linked skills explicitly',
     'cannot handle the task.',
   ]