Skill · v1.0
Batch Archival
Batch Archival moves finished work into your archive in batches, under per-artifact approval gates. The scan proposes candidates by terminal status, a readiness audit checks each one will land clean and nothing moves until you approve it. Completed work leaves the active workspace without a single accidental loss.
Install
Claude Code (CLI / WSL / Git Bash)
/plugin marketplace add https://www.infinitegameos.io/marketplace.json
/plugin install batch-archival@igos-libraryClaude Code (VS Code)
Install in VS CodeOpens the Claude Code plugins dialog with the marketplace and skill prefilled. Requires the Claude Code VS Code extension installed and signed in. Or paste the snippet below into .claude/settings.json for VS Code, JetBrains or any setup that prefers manual config.
{
"extraKnownMarketplaces": {
"igos-library": {
"source": {
"source": "url",
"url": "https://www.infinitegameos.io/marketplace.json"
}
}
},
"enabledPlugins": {
"batch-archival@igos-library": true
}
}Direct markdown URL (Claude Code, Cursor, Codex CLI)
https://www.infinitegameos.io/markdown/skills/batch-archivalThis URL returns the narrative concept page. The plugin install path above delivers the operational SKILL.md instruction file.
Cursor (.mdc rules file)
curl -O https://www.infinitegameos.io/install/cursor/batch-archival.mdcAider, Cline, any agent with --read
curl -O https://www.infinitegameos.io/markdown/skills/batch-archival
aider --read batch-archival.mdThis skill ships in both homes, the Sovereign Ecosystem Foundation and this library.
Definition
Batch Archival is a skill for archiving several workspace artifacts at once with approval gates at every step that matters. An active workspace accumulates finished work: implemented plans, completed notes, retired projects. Left in place, they blur the line between what is alive and what is done. Moved carelessly, a batch sweep breaks links and buries something still needed. Batch Archival threads that needle. It scans your active containers for artifacts carrying a terminal status, classifies them by domain, runs a lightweight readiness audit on each candidate (metadata completeness, links, destination routing) and presents an approval list. The operator approves per artifact or per domain. Only then do files move. Status never changes as a side effect: archival is a location change and the lifecycle status stays what it was. Draft, proposed and in-flight work is excluded by default because those statuses mean the work is still alive.
Terminal status is the eligibility signal
A file becomes an archival candidate because its lifecycle ended, not because it looks old. Plans qualify at implemented. Non-plan artifacts qualify at complete. Object types with their own vocabulary qualify at their own terminal status, like retired for quest-style notes. Everything still carrying draft, proposed, approved or ready-for-execution is excluded by default, because those statuses mean the work is alive. Age is not a status. A three-month-old plan that never shipped is stalled, not archivable, and it needs a different conversation.
Where an object type distinguishes completed from retired, the metadata preserves the distinction even when both land in a shared archive folder. The archive stays flat and the frontmatter stays honest, which keeps future review workflows possible without folder sprawl.
The readiness audit and the approval gate
Before anything is proposed for a move, a lightweight audit checks that the move will land clean: the destination path resolves, domain routing is correct, object-specific links hold and any support-file subtrees a plan kept during execution are dispersed to their permanent homes first. The audit is proposal-only. It produces notes and blockers, never moves.
The approval gate is the spine. Early operation requires explicit approval per domain or per artifact, and one pilot batch runs before full activation. Time-gated material like transcripts past a holding period gets its own proposal-only audit. Automation is earned one proven artifact class at a time, never assumed. The scan proposes, the operator disposes.
Use Cases
Quarterly sweep of implemented plans
A quarter of steady shipping leaves a dozen implemented plans sitting in the active planning folder. Batch Archival scans on terminal status, groups the candidates, audits that each plan's support files have been dispersed and presents the approval list. One approval per domain and the planning folder returns to holding only what is actually in motion.
The stale-notes cleanup that finally happens
Completed project notes, finished checklists and retired experiments have accumulated across three containers. A manual cleanup keeps getting deferred because moving files one at a time is tedious and moving them in bulk feels dangerous. Batch Archival makes the bulk move safe: every candidate is audited and listed, and nothing leaves until approved. The cleanup that was always postponed takes one session.
Catching the artifact that was not ready
A sweep proposes eleven candidates. The readiness audit flags one: a completed creative artifact whose companion strategy note never got its backlink. The flag surfaces as a blocker in the approval list rather than a broken link discovered months later. The operator fixes the link, approves the batch and the archive stays coherent.
FAQ
Why per-artifact approval instead of trusting the status scan?
Because a status field can be wrong, stale or ambiguous, and an archive move buries the mistake. The scan is the intelligence pass and the approval is the decision. Early on, every artifact or domain gets explicit approval. After stable operation, automation can open one proven artifact class at a time. The gate loosens with evidence, never with optimism.
Does archiving change an artifact's status?
No. Archival is a location change and nothing else. An implemented plan stays implemented, a retired note stays retired. Keeping status and location independent means the archive remains queryable by lifecycle state, and it removes a whole class of accidental status regressions from bulk operations.
What about material that becomes archivable on a clock rather than a status?
Time-gated material like transcripts or session notes past a holding period gets a proposal-only eligibility audit. The audit surfaces candidates with their gate condition, and the operator approves before anything moves. The clock qualifies a candidate; it never executes a move.
Related
Skill
Session Closeout
A fast three-action close for a git-tracked workspace. Breadcrumbs as you work, a refreshed primer, a readable commit as the session summary.
Skill
Pending Plan Implementation
Execute approved plans with checkpoint gates, breadcrumb logging and outcome summaries.
Skill
Autonomous Improvement Session
Run a set-and-forget session that improves your system without active attention. Operator-invoked only. Safe, additive hygiene work executes directly; judgment forks and floor-class items surface for your review.
Batch Archival is the sweep that keeps the active workspace honest. Session Closeout marks the work that queues for it, and Pending Plan Implementation produces the implemented plans it eventually carries home. All three ship inside the Sovereign Ecosystem Foundation. The design frame for a workspace worth keeping clean is the Sovereign Life Playbook.
See the Sovereign Life Playbook