Skip to main content
Exécutez n'importe quel Skill dans Manus
en un clic
diazMelgarejo
GitHub creator profile

diazMelgarejo

Repository-level view of 36 collected skills across 2 GitHub repositories, including approximate occupation coverage.

skills collected
36
repositories
2
occupation fields
3
updated
2026-06-03
repository explorer

Repositories and representative skills

#001
orama-system
28 skills00updated 2026-06-03
78% of creator
orama-system
non classé

Elegant problem-solving methodology with 5-stage process, AFRP pre-router gate, CIDF v1.2 content insertion framework, and 7-agent execution network. Activates for architectural thinking, systematic verification, content insertion decisions, complex multi-step tasks, code quality reviews, and self-improvement workflows. Triggers on: "ultrathink", "think deeply", "5-stage", "systematic approach", "elegant solution", "verify before done", "content insertion", "AFRP", "CIDF".

2026-06-03
git-reanchor
non classé

Recover branches that look "orphaned" or "600 commits behind" after a force-pushed / rewritten main, by re-anchoring each onto the byte-identical twin commit that already exists in the rewritten history. Invoke when: "branches are 600 behind", "orphaned branch after rewrite", "re-anchor branch to main", "branches lost common ancestor", "reconcile branches after force-push", "recover deleted branch", "byte-identical common ancestor", "git history rewrite recovery", "branches all became the same".

2026-06-03
gstack
non classé

gstack v1.12.2.0 integration sub-skill. Full routing table for web browsing, QA, shipping, planning reviews, design, DX audits, retros, and GBrain. Activates for: /browse, /qa, /ship, /review, /investigate, /design-review, /canary, /benchmark, /retro, gbrain, gstack skills, web browsing, QA testing, deploy, design review, canary monitoring, performance benchmarks.

2026-06-02
expunge-git
non classé

Complete git-history scrub procedure for removing forbidden tokens, leaked secrets, or contaminated author identities from a repo's commits, branches, reflog, and packed objects. Invoke when: "expunge git history", "remove from history", "rewrite author", "scrub commits", "delete from reflog", "clean leaked secret from git", "remove forbidden token", "rewrite history", "purge identity string".

2026-06-02
mcp-install
non classé

Use when setting up the MCP orchestration stack on a new machine or verifying it is installed. Covers ai-cli-mcp (parallel background workers), the OpenClaw MCP registry, and an optional Gemini analyzer lane when explicitly requested. Activates for: install mcp stack, setup gemini mcp, register ai-cli, openclaw mcp set, mcp orchestration setup, install mcp tools, run install-mcp-stack.sh.

2026-06-02
mcp-orchestration
non classé

Use when setting up Claude Code MCP tools, adding Claude Skills, using Gemini as a large-context reader, dispatching parallel AI CLI agents, connecting MCP tools to OpenClaw, or routing tasks to local/OpenRouter/Codex/Gemini agents. Trigger when the user mentions Gemini MCP, ai-cli-mcp, OpenClaw MCP, SKILL.md, Claude Skills, background agents, /mcp, MCP JSON, prompt caching, OpenRouter routing, agent dispatch, or tool setup failures.

2026-06-02
no-sleep-chains
Développeurs de logiciels

Enforced rule for all agents: sleep N && <command> chains are blocked by the shell hook. Invoke this skill whenever you need to wait for a background process, a file to grow, or a condition to become true. Covers: background task output polling, npm install completion, claude update, any long-running subprocess.

2026-06-02
skillify
Développeurs de logiciels

Interactive skill creator for orama-system, raw Claude Code, and gstack. Creates a new skill folder with SKILL.md, frontmatter, boundaries, examples, and harness registration — by asking you questions, then writing the files. Activates when asked to: create a skill, new skill, add a sub-skill, /skillify, build a new tool as a skill, make a skill, install a skill. Supports targets: raw-claude, orama-system sub-skill, gstack global skill, all.

2026-06-02
Showing top 8 of 28 collected skills in this repository.
#002
Perpetua-Tools
8 skills10updated 2026-05-24
22% of creator
alphaclaw-session
Développeurs de logiciels

Commandeer AlphaClaw/OpenClaw runtime defaults. Set environment profiles, backup/restore sessions, enumerate live agents, self-heal connectivity issues. Run when starting any OpenClaw-dependent session.

2026-05-24
startup-intelligence
Développeurs de logiciels

Startup scenario classification, probe retry, history-driven adaptive timeouts, cloud fallback, and .env.local override detection for Perpetua-Tools and orama-system. Use when startup fails to reach backends, when you need to understand which scenario the system is in (FULL_DISTRIBUTED, MAC_DUAL, MAC_OLLAMA_ONLY, MAC_LMS_ONLY, CLOUD_ONLY, FULLY_OFFLINE), or when routing behaves unexpectedly after reboot.

2026-05-20
perpetua-tools-conventions
Développeurs de logiciels

Development conventions and patterns for Perpetua-Tools. Python project with mixed commits.

2026-05-07
self-discovery
Administrateurs de réseaux et de systèmes informatiques

Query live state across all 3 repos — versions, branch status, LM Studio endpoints, file manifests. Run whenever you need a situational snapshot of the stack.

2026-04-29
self-improve
Développeurs de logiciels

Crystallize session learnings into LESSONS.md and instincts. Runs automatically at session end (Option C — proposes updates, commits only after user approval). Also invocable on-demand.

2026-04-26
perpetua-tools-conventions
Développeurs de logiciels

Development conventions and patterns for Perpetua-Tools. Python project with mixed commits.

2026-04-25
model-routing-check
Administrateurs de réseaux et de systèmes informatiques

Verify LM Studio endpoint reachability and routing table validity before any agent dispatch

2026-04-20
agent-run
Administrateurs de réseaux et de systèmes informatiques

Launch a Perpetua-Tools agent with LM Studio endpoint validation and env setup

2026-04-20
2 sur 2 depots affiches
Tous les depots sont affiches