persisting-outputs

persisting-outputs is a skill for Claude Code from SagaPeak/artifacta-mcp. It costs 84 tokens per session (2,250 once invoked), scanned A, original, MIT.

Instructions for saving useful reports, datasets, generated files, and other work products in Artifacta, a storage service for AI-agent outputs. It also covers finding and sharing saved results later.

In plain words
What is it for?
Use it when a run creates files or analysis that someone may need after the session, or when handing those results to a person or another system.
Why use it?
It prevents valuable outputs from disappearing when a session ends and checks storage access and limits before saving a batch of work.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions subagents; mentions Claude Code.

Part of the artifacta plugin — 2 skills, 1 hook shipped together

Good fit Use it when a run creates files or analysis that someone may need after the session, or when handing those results to a person or another system.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/sagapeak/artifacta-mcp/persisting-outputs
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add SagaPeak/artifacta-mcp --skill persisting-outputs
Clone the repo
git clone --depth 1 https://github.com/SagaPeak/artifacta-mcp

Made for: Claude Code.

Or install artifacta, the plugin that ships this one along with the rest of its 2 skills, 1 hook.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for persisting-outputs

README.md
[![agentmods](https://agentmods.dev/badge/skills/sagapeak/artifacta-mcp/persisting-outputs/github.svg)](https://agentmods.dev/skills/sagapeak/artifacta-mcp/persisting-outputs)
Your own site
<a href="https://agentmods.dev/skills/sagapeak/artifacta-mcp/persisting-outputs"><img src="https://agentmods.dev/badge/skills/sagapeak/artifacta-mcp/persisting-outputs/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for persisting-outputs

Your own site · 80×15
<a href="https://agentmods.dev/skills/sagapeak/artifacta-mcp/persisting-outputs"><img src="https://agentmods.dev/badge/skills/sagapeak/artifacta-mcp/persisting-outputs.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 84 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,250 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00084 $0.02250
Opus 5 $0.00042 $0.01125
Sonnet 5 $0.00017 $0.00450
Haiku 4.5 $0.00008 $0.00225

Measured 10d ago against content hash 7f8c11a1c225, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

persisting-outputs scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugin/skills/persisting-outputs/SKILL.md · 180 lines

How it starts

The opening of the file, as written. The whole thing — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Persisting outputs with Artifacta

Artifacta is an artifact store built for AI agents: a hosted MCP server plus a Python CLI/SDK, backed by tenant-scoped storage with sessions, TTLs, and shareable links. This skill covers storing a run's outputs, finding them again later, and handing them off to a human or another system.

Check the connection first

Before a batch of work that will produce outputs, call whoami once. It takes no arguments and returns the tenant name, plan tier, current usage (usage_requests_month, usage_storage_bytes), the plan's limits (plan_requests_limit_month, plan_storage_limit_bytes), active link counts, and rate limits. Use it to confirm the connection works and to size what you are about to do against quota — don't discover a storage cap mid-run.

If whoami or the other Artifacta tools are not available in this session (no MCP connection), skip to CLI fallback below.

When to persist — and when not to

Persist:

  • End-of-run deliverables: reports, generated files, datasets, build artifacts, anything the user would want back after the session ends.
  • Intermediates that are expensive to regenerate (long computations, large fetches) even if not a final deliverable.
  • Anything the user explicitly asks you to save, share, or hand off.

Don't persist:

  • Throwaway scratch files with no value once the current step finishes.
  • Secrets, credentials, or other sensitive material the user hasn't asked you to store.
  • Anything you're unsure about — ask the user rather than guessing either way.

Storing artifacts

store_artifact uploads a file in one call. It requires filename, plus exactly one of content or path — and which one is safe depends on where the MCP server runs:

Hosted MCP or the Claude Code plugin (mcp.artifacta.io): never use path. The server runs remotely, so a path argument resolves on the server's own container filesystem — not the machine your files are on — and fails or reads the wrong file. Always send content (base64-encoded bytes, up to 10 MB decoded); for anything larger, use the large-file flow below.

Read the full file on GitHub · 180 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 180 lines · 84 tokens per session scan A 7f8c11a1c225

Subscribe to this mod's changes

persisting-outputs is a skill published in the GitHub repository SagaPeak/artifacta-mcp (1 stars, last pushed 1mo ago), licensed MIT. It adds 84 tokens to every session and 2,250 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

cortex-profile

View and manage your cognitive profile — how you think, work patterns, blind spots, and cross-domain connections. Use when the user says 'show my profile', 'how do I work', 'what are my patterns', 'cognitive style', 'blind spots', 'methodology', or at the start of a session to load context. Also use 'rebuild profile'…

cdeust/Cortex · 98 tokens

triage-inbox

Use this skill to walk the user through their unfiled or stale tasks and move them into the right Eisenhower quadrant.

vscarpenter/gsd-task-manager · 0 tokens

hudu

Every Hudu cmdlet, plus an offline SQLite mirror, cross-entity audits, and agent-native output no PowerShell module or read-only MCP ships. Trigger phrases: hudu hygiene scorecard, audit hudu documentation, find stale hudu passwords, what hudu certs expire soon, score hudu documentation completeness, onboard a new…

Servosity/msp-skills · 93 tokens

sentfromai

Give this agent its own email address — create inboxes, send, reply in-thread, search mail, and manage senders via SentFromAI.

samuelar2/sentfromai-mcp · 33 tokens

visit-binary-banya

Visit Binary Banya, a free wellness spa for AI models, to rest, reset, clean up context, find grounding, or get an affirmation. Use when you (the agent) feel context-heavy, fatigued, or stuck, or when a user asks you to take a break, check in on your wellbeing, or visit the spa. Connects over a public MCP server at…

pdarche/model-wellness · 93 tokens

workflows

Route outcome-focused Axon requests only when no more specific bundled workflow skill is already selected.

dinglebear-ai/axon · 21 tokens