What ran

Daily skills-scheduled-mirror-sync. Verifies that canonical SKILL.md files under Projects/Skills/ match the loader paths under ~/Documents/Claude/Scheduled/ that scheduled-task prompts actually pull from.

Pre-flight

Mirror sync results — no drift

Canonical Destination SHA256 Action
Skills/post-run-qa-reflection/SKILL.md Claude/Scheduled/_post-run-qa-reflection.md ecf9a738… no-op (in sync)
Skills/sail-kb-daily-pulse/SKILL.md Claude/Scheduled/sail-kb-daily-pulse/SKILL.md a0f221c2… no-op (in sync)
Skills/sail-kb-daily-pulse/build_report.py Claude/Scheduled/sail-kb-daily-pulse/build_report.py 69e5129a… no-op (in sync)

Drift count: 0. Files written: 0. Commits made: 0.

Skills_Library compatibility (informational)

Cross-tool skill surface check

Ran routines-build/scripts/skill_surface_check.py --markdown. No flags.

Loader/mirror surfaces (counts): Projects/Skills 116, Skills_Library 0, Claude Scheduled 60, Claude Skills 5, Codex ~/.codex/skills 10. Cursor managed skills, Codex skills, and Claude custom skills (app support) absent — expected.

Editor/MCP surfaces: Projects .mcp.json present, .claude/settings.local.json present, Cursor MCP present, VS Code user settings present. Repos/.mcp.json absent — expected.

API/model surfaces (manifest-backed): ChatGPT/OpenAI, Anthropic, Perplexity, Gemini, Kimi, OpenRouter, Manus, Cursor, Windows MCP — all covered via AGENT-TOOL-MANIFEST.

Same-slug overlap (info only): 6 slugs in both Projects/Skills and Claude/Scheduled. 110 only in Projects/Skills. 54 only in Claude/Scheduled.

Artifacts