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.
npx skills add Sistema2D/FrameCode-VibeWork --skill wiki-curatorgit clone --depth 1 https://github.com/Sistema2D/FrameCode-VibeWorkWrote 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.
[](https://agentmods.dev/skills/sistema2d/framecode-vibework/wiki-curator)<a href="https://agentmods.dev/skills/sistema2d/framecode-vibework/wiki-curator"><img src="https://agentmods.dev/badge/skills/sistema2d/framecode-vibework/wiki-curator/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.
<a href="https://agentmods.dev/skills/sistema2d/framecode-vibework/wiki-curator"><img src="https://agentmods.dev/badge/skills/sistema2d/framecode-vibework/wiki-curator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 39 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00040 | $0.01954 |
| Opus 5.5 | $0.00016 | $0.00782 |
| Sonnet 5.5 | $0.00008 | $0.00391 |
| Haiku 4.5 | $0.00004 | $0.00195 |
Grade A, and why
wiki-curator 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 215 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SKILL: Wiki Curator
Maintain the LLM Wiki as a continuously improving knowledge base while preserving FCVW's pillars: updates on demand, measurable freshness, and low token cost.
Purpose
Promote sourced reusable knowledge into the smallest canonical wiki surface without duplicating records or loading the full archive by default.
Activation Triggers
Load this skill when a task involves:
- creating, updating, merging, or reviewing wiki knowledge;
- promoting plans, changelogs, troubleshooting, audits, decisions, or sessions into reusable wiki pages;
- grouping related notes under the same topic;
- organizing tags, themes, or frontmatter colors;
- checking stale, duplicated, contradictory, or superseded wiki content;
- preparing minor or major releases that add reusable knowledge.
Inputs
Changed source records, current canonical page candidates, wiki/README.md, taxonomy, metrics, active index/log, typed relationship findings, stale-source review candidates, and an explicit source-bounded search scope.
Fixed Cost Mode
Always use the standard optimized mode. Do not ask the user to choose a cost mode.
Load only:
AGENTS.mdchecklist summary.CONTEXT_MAP.mdrow for the active session type.wiki/index.mdandwiki/README.md.- the taxonomy in
wiki/README.mdwhen theme, tag, or freshness decisions are needed. - Source files that directly triggered the curation: changed plans, changelogs, troubleshooting records, audits, decisions, sessions, or specific wiki pages.
Do not read every wiki page unless governance-validator (wiki lint mode) reports an anomaly that requires it.
Curation Loop
- Collect trigger sources: identify new or changed source records and load only their summaries or relevant sections.
- Decide promotion: promote only knowledge that meets
wiki/README.mdpromotion criteria. - Merge before create: update an existing related page when the topic already exists; create a new page only for a distinct reusable concept.
- Cluster topics: link related pages with Obsidian-style links and record canonical pages for superseded or overlapping notes.
- Refresh metadata: set
last_reviewed,related_version,tags,theme,theme_color,maturity, andnext_reviewwhen applicable; do not require maturity for source, raw, or session records. - Use typed relations deliberately: prefer the narrowest meaningful relation, resolve every target, and rely on generated inverse edges instead of copying both directions.
- Review changed sources: when a tracked
source_digestchanges, inspect only pages linked throughderived_from; confirm, update, supersede, or invalidate them before refreshing the digest. - Update metrics: record freshness, promotion, duplication, source-impact, and taxonomy results in the active plan.
- Log curation: summarize the curation in the active plan or session note; git history is the log.
- Validate: run deterministic
governance-validator(wiki lint mode) for minor/major releases or when three or more wiki pages changed; semantic review remains optional and source-bounded.
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.
- 1mo ago Changed · +105 lines · +27 tokens per session 55f64e0fdbdb
- 1mo ago Changed · +5 lines 1059823414d0
- 2mo ago First seen · 105 lines · 13 tokens per session scan A c82cc5ce6d4a
wiki-curator is a skill published in the GitHub repository Sistema2D/FrameCode-VibeWork (29 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 40 tokens to every session and 1,954 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-08-30.
wiki-curator compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| wiki-curatorSistema2D/FrameCode-VibeWork | Skill | 29 | 1mo ago | Apache-2.0 | A | 40 | Claude Code, Codex |
| rememberyonatangross/orchestkit | Skill | 299 | yesterday | MIT | A | 66 | Claude Code |
| memoryyonatangross/orchestkit | Skill | 299 | yesterday | MIT | A | 36 | Claude Code |
| sage-memoryl33tdawg/sage | Skill | 262 | yesterday | Apache-2.0 | A | 115 | Claude Code, Codex |
| wenlan-handoffwenlan-app/wenlan | Skill | 93 | yesterday | Apache-2.0 | A | 43 | Claude Code, Codex |
| curatewenlan-app/wenlan | Skill | 93 | yesterday | Apache-2.0 | A | 106 | Claude Code, Codex |
| wenlan-recallwenlan-app/wenlan | Skill | 93 | yesterday | Apache-2.0 | A | 54 | Claude Code, Codex |