feat: Split bot capabilities into individual skill files under .github/skills/

[Copilot]

SKILL.md did not exist in the repo; the request was to document each bot capability as a separate, detailed skill file in .github/skills/.

Changes

• SKILL.md (root) — lightweight index table linking to all skill files
• .github/skills/node-monitoring.md — /node: JSON-RPC interrogation, extract_address_data, balance snapshot with threading lock, validation chart lifecycle
• .github/skills/balance-history.md — /hist + /flush: JSON persistence format, filter_last_24h/filter_since_midnight, chart generation, ConversationHandler confirmation flows, legacy key format compatibility
• .github/skills/scheduled-reports.md — APScheduler every 60 min, async/sync bridge (run_coroutine_threadsafe vs run_until_complete), report logic at 7h/12h/21h, balance delta calculation, node-down detection criteria
• .github/skills/crypto-prices.md — /btc via API-Ninjas (auth header), /mas via two parallel MEXC endpoints (asyncio.gather), HTTP retry wrapper
• .github/skills/system-monitoring.md — /hi (git hash), /temperature (psutil, Linux-only sensors), /perf (RPC latency + uptime from history density), legacy timestamp parsing
• .github/skills/docker-management.md — full /docker ConversationHandler state machine, Docker SDK socket usage, exec_massa_client flow (buy/sell rolls input → confirm → exec), why @auth_required is not used on entry points
• .github/skills/authentication.md — @auth_required vs @cb_auth_required (callback queries return ConversationHandler.END), manual auth pattern for ConversationHandler entry points, whitelist loading from topology.json

Original prompt

sépare ce fichier SKILL.md en plusieurs skills que tu mets dans .github/skills/.
Détaille chaque sous skill

The user has attached the following file paths as relevant context:

• SKILL.md

Created from VS Code.

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.