Menú
Pull generalized contributions from a project into Memnyx. Runs `sanitizer --check` as a mandatory gate before integration.
Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.
Basado en la clasificación ocupacional SOC
| name | pull-contributions |
| description | Pull generalized contributions from a project into Memnyx. Runs `sanitizer --check` as a mandatory gate before integration. |
| user_invocable | true |
| args | Path to a workspace's contributions/ folder (e.g., <workspace>/contributions/) |
See .claude/skills/_shared/MODEL_SELECTION.md (in your workspace) for full policy.
You are integrating generalized contributions from a project into Memnyx. This skill is meant to be run from the Memnyx repo, not from a project.
Locate contributions: The user provides the path to <workspace>/contributions/ (e.g., ~/workspace/contributions/). Read all .md files in that folder that do not start with integrated- or skipped-. Files with the integrated- prefix have already been applied; files with the skipped- prefix were reviewed and rejected — skip both silently. sanitizer-report-*.md files are scan reports, not contributions — skip those too.
If no path is provided, ask the user for it.
Sanitize gate (MANDATORY): Run sanitizer in check mode on the contributions folder:
/sanitizer <contributions-folder-path> --check --mode=boilerplate
0 → proceed.1 → halt. Surface the sanitizer report. Do NOT pull any contribution with findings. Ask the user to run /sanitizer <path> --apply in the source project to fix, then re-run this skill.Review each contribution (content check only, not PII/secret — that's sanitizer's job):
Present a summary of all contributions found:
Contributions found: [count]
─────────────────────────────
1. [title] (type: skill-update, target: .claude/skills/hello/SKILL.md)
2. [title] (type: convention, target: CLAUDE.md)
...
Ask: "Which contributions should I integrate? (all / specific numbers / none)"
For each approved contribution, apply the changes:
.claude/docs/conventions.md. Budget gate for always-loaded targets: if the target is a file CLAUDE.md @-imports (e.g. .claude/docs/agent-guardrails.md), measure it first (tokens ≈ bytes/4) against the always-loaded thresholds in .claude/skills/finance-controller/references/thresholds.md. At yellow (>2,000 tokens) or above, the addition is zero-sum: draft a displacement or condensation of at least the addition's size in the same change, and let the user decide which rule loses its slot. Safety rules (production read-only, secrets handling, and similar) are exempt from displacement and may land at yellow — flag the overage instead, so compression rebalances among style and process rules..mcp.json and settings.json.Always ask for approval before writing any changes. Show exact diffs or new content.
After integration, ask: "Should I mark the processed contributions as integrated?"
integrated- prefix (e.g., integrated-2025-06-15-improve-hello-mcp-retry.md), and contributions the user reviewed and rejected to add a skipped- prefix. Both prevent re-processing on future runs.Update CLAUDE.md if new skills were added (add to the skills table).
Report:
Pull Complete
=============
Reviewed: [count] contributions
Integrated: [count]
Skipped: [count] (with reasons)
Files modified: [list]
Remember to commit the changes to the Memnyx repo.
Authoritative, version-pinned reference for every Claude Code CLI flag, classified by whether it works in an INTERACTIVE session (default TUI), a HEADLESS session (-p/--print), or BOTH — plus the canonical launch recipes for each mode. Use when any session needs to launch or script `claude` and must know which switch is valid in which mode, build a headless/streaming invocation, set up remote-control or an always-on session, pass model/effort/permissions/MCP/plugin/settings flags, or answer "does flag X work with -p / in interactive". Trigger: "claude flags", "claude cli switches", "which flags work headless vs interactive", "-p flags", "stream-json flags", "how do I launch claude with X", "/claude-cli-flags".
Authoritative, version-pinned reference for every built-in Claude Code slash (/) command — what each does, its aliases, and which are bundled Skills/Workflows vs fixed built-ins — plus how custom slash commands work (the commands-merged-into-skills model, file locations, frontmatter, arguments, !bash, @file). Includes the full /loop vs /schedule breakdown. Use when any session needs to know what a slash command does, find the right command for a task (clear context, switch model, review code, schedule work, background a task), or author a custom /command. Trigger: "claude slash commands", "what does /X do", "list the slash commands", "built-in commands", "custom slash command", "/loop vs /schedule", "how do I make a /command", "/claude-slash-commands". Sibling of claude-cli-flags (launch flags).
Initialise a workspace, register projects under it, or sync upstream changes. Single entry point for workspace lifecycle.
Enforce coordinated integration testing — identify tool dependencies, check running systems, test all 4 levels (unit/component/integration/system), never disrupt workflows
Autonomous task loop — runs a task repeatedly until completion criteria are met. No babysitting required.
Dual-model review — runs the same question past a second, independent reasoner (a separate model/CLI, e.g. Codex) and synthesizes the disagreements with the primary Claude take. Use before an irreversible call — architecture, security, strategy, high-blast-radius code. Trigger phrases "second opinion", "/second-opinion", "review this", "get another perspective", "sanity check this plan".