Skip to main content
skill CX Codex

check-skill-provenance-stamps-before-archiving-knowledge-entries skill

check-skill-provenance-stamps-before-archiving-knowledge-entries is a Claude Code skill that checks source_refs and UUIDs before archiving knowledge entries, stopping removal if an active skill depends on one.

MIT ★ 539 updated 2026-10-29

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add ZaxbyHub/opencode-swarm --skill check-skill-provenance-stamps-before-archiving-knowledge-entries -a claude-code
Open on GitHub

What check-skill-provenance-stamps-before-archiving-knowledge-entries does

This skill runs before you archive an entry in the knowledge base. The agent compares source_refs inside .opencode/skills/generated/*/SKILL.md against the entries on your archive list, and stops when an active skill still depends on one of them.

Why it matters: archiving a source entry leaves a stale marker behind, and the generated skill quietly drops out of injection. The check is a grep for the entry UUID across generated skills. A reviewer then confirms nothing depends on it.

The skill carries written steps and a delegation line of the form SKILLS: file:.opencode/skills/generated/check-skill-provenance-stamps-before-archiving-knowledge-entries/SKILL.md. opencode-swarm's generator built it from a single knowledge entry, so the .opencode/skills/generated/ path is baked into the procedure: point it at your own tree first, otherwise the grep comes back empty for the wrong reason.

What it helps with

  • confirm a knowledge entry is safe to archive
  • find every generated skill that references an entry UUID
  • match source_refs in SKILL.md against the archive list
  • hand an agent the skill path in a SKILLS line
  • explain to a reviewer why an entry has to stay active

Example prompts for the agent

  • Check the provenance stamps before archiving entry f80f0942-3236-4b5d-9b98-a23578676f27.
  • Grep .opencode/skills/generated/*/SKILL.md and tell me which skills depend on this entry.
  • Is it safe to archive this entry while an active skill still references it?

How to install check-skill-provenance-stamps-before-archiving-knowledge-entries

1Skills CLI (Vercel)

npx skills add ZaxbyHub/opencode-swarm --skill check-skill-provenance-stamps-before-archiving-knowledge-entries -a claude-code

2Manually: put the files in place

~/.claude/skills/check-skill-provenance-stamps-before-archiving-knowledge-entries/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/check-skill-provenance-stamps-before-archiving-knowledge-entries/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about check-skill-provenance-stamps-before-archiving-knowledge-entries

How do I install check-skill-provenance-stamps-before-archiving-knowledge-entries in Claude Code?

Run npx skills add ZaxbyHub/opencode-swarm --skill check-skill-provenance-stamps-before-archiving-knowledge-entries -a claude-code. The skill file lands at ~/.claude/skills/check-skill-provenance-stamps-before-archiving-knowledge-entries/SKILL.md.

What does check-skill-provenance-stamps-before-archiving-knowledge-entries do?

It has the agent check generated skills' source_refs against the entries you plan to archive, and hold off when an active skill still depends on one of them.

What happens if I archive a source entry anyway?

A stale marker appears and the generated skill is silently excluded from injection, so the agent stops seeing it. Checking the UUID beforehand is what avoids that.

Source text

SKILL.md · MIT
---
name: check-skill-provenance-stamps-before-archiving-knowledge-entries
description: check skill provenance stamps before archiving knowledge entries
metadata:
  generated_from_knowledge: '["f80f0942-3236-4b5d-9b98-a23578676f27"]'
  source_knowledge_ids: '["f80f0942-3236-4b5d-9b98-a23578676f27"]'
  generated_at: "2026-07-14T05:31:49.340Z"
  confidence: "0.60"
  status: active
  version: "1"
  skill_origin: generated
---

<!-- generated by opencode-swarm skill-generator. Do not edit by hand; edits will be preserved on regeneration only with controlled update mode. -->

# check skill provenance stamps before archiving knowledge entries

## Trigger

- (no explicit trigger metadata; cluster derived from category/tags)

## Required Procedure

- check skill provenance stamps before archiving knowledge entries
- verify .opencode/skills/generated/*/SKILL.md source_refs against entries being archived

## Forbidden Shortcuts

- (none recorded)

## Delegation Template

When delegating a task affected by this skill, include:

```
SKILLS: file:.opencode/skills/generated/check-skill-provenance-stamps-before-archiving-knowledge-entries/SKILL.md
```

## Reviewer Checks

- grep generated skills for the entry UUID being archived
- confirm no active skill depends on entry before archive

## Source Knowledge IDs

- f80f0942-3236-4b5d-9b98-a23578676f27 — Before archiving knowledge entries, verify they are not source entries for active generated skills — archiving source entries triggers stale markers that silently exclude skills from injection.

© ZaxbyHub · MIT · ZaxbyHub/opencode-swarm

Similar skills

All skills for Agent setup: memory, context, orchestration →

From the same set121

The whole set →
run-relevant-tests-after-adding-error-handling-wrappers run-relevant-tests-after-adding-error-handling-wrappers is a Claude Code skill that prompts an agent to run relevant tests after adding error handling and check the success path's original return type and value. CX Codex +1 ★ 539repo stars issue-tracer issue-tracer walks a bug from reproduction to a PR-ready fix: locate the cause, plan the patch, get it reviewed. CX Codex +1 ★ 539repo stars swarm-ci-monitor swarm-ci-monitor monitors CI runs on approved pull requests, repairs broken checks within five cycles, and merges the branch once tests pass. CX Codex +1 ★ 539repo stars restore-internals-seams-in-finally-blocks-after-each-test restore-internals-seams-in-finally-blocks-after-each-test is a Claude Code skill that directs agents to restore original _internals in finally after each test of error paths through DI seams. CX Codex +1 ★ 539repo stars consult consult enforces an advisory protocol with strict read-before-cite checks and explicit uncertainty levels. CX Codex +1 ★ 539repo stars run-affected-test-suite-end-to-end-after-multi-file-signature-ch run-affected-test-suite-end-to-end-after-multi-file-signature-ch is a Claude Code skill that runs affected tests and checks every call site after function signatures change across files. CX Codex +1 ★ 539repo stars issue-tracer issue-tracer takes a bug from reproduction to a PR-ready fix: locates the cause, plans the change, and puts it through critic review. CX Codex +1 ★ 539repo stars confirm-failures-are-causally-linked-to-the-task-before-reportin confirm-failures-are-causally-linked-to-the-task-before-reportin is a Claude Code skill that separates pre-existing test failures from new regressions before rejecting work. CX Codex +1 ★ 539repo stars

Project sponsors

Companies that keep this analytics open