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 kerbelp/metatron --skill context-okf-promote-candidatesgit clone --depth 1 https://github.com/kerbelp/metatronWrote 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/kerbelp/metatron/context-okf-promote-candidates)<a href="https://agentmods.dev/skills/kerbelp/metatron/context-okf-promote-candidates"><img src="https://agentmods.dev/badge/skills/kerbelp/metatron/context-okf-promote-candidates/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/kerbelp/metatron/context-okf-promote-candidates"><img src="https://agentmods.dev/badge/skills/kerbelp/metatron/context-okf-promote-candidates.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00056 | $0.01259 |
| Opus 5 | $0.00028 | $0.00629 |
| Sonnet 5 | $0.00011 | $0.00252 |
| Haiku 4.5 | $0.00006 | $0.00126 |
Grade A, and why
context-okf-promote-candidates 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 11d 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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Promoting candidates to canonical (OKF, git-as-truth)
Overview
A Metatron decision becomes canonical by moving its OKF file from
context/candidate/ to context/decisions/. The directory is the status.
This skill covers doing that move in a git/CI flow, as a companion to authoring
candidates (see the context-okf-llm-ingest skill).
This skill is mechanical only. You move the specific files a human named, and nothing else. You do not decide which candidates are worthy — that judgment is the human curation act Metatron's model is built around.
Directory name:
context/is the default knowledge-base directory. A repo may configure another name (context_dirinmetatron.toml,METATRON_CONTEXT_DIR, or--context-diron the mirror commands); pre-rename repos may still usemetatron/. The layout inside is identical — substitute the configured name.
The invariant — read before doing anything
Crossing the canonical boundary is always human-gated. Nothing self-promotes.
In a git flow, the gate is a human approving the specific pull request that
contains the candidate/ → decisions/ move. That approval is the curation act.
So, absolutely:
- Only move candidates a human explicitly named (by id/slug, or as the agreed contents of the promotion PR). Never sweep "these look good."
- Never auto-merge. Do not promote on green CI, on a schedule, or via a bot that merges its own PR. CI may validate and rebuild the serving index; the merge is a human's.
- Promotion is a pure move. Do not edit
pattern/rationale/scopein the same change — editing content is authoring, a separate act with its own review. - One promotion PR, kept legible. Don't bury promotions inside a large authoring diff; a reviewer must be able to see exactly what is crossing the boundary.
If you are tempted to choose what to promote, stop — that is the human's job.
Model: git is the source of truth
Files stay id-less; the directory (candidate/ vs decisions/) is the status.
metatron mirror import rebuilds the serving database from the files — the DB is a
derived, rebuildable index, not something you curate or hand-edit. Do not stamp
id fields into files (an id the rebuilt DB doesn't know is skipped on import).
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.
- 11d ago First seen · 102 lines · 56 tokens per session scan A 66040af008d0
context-okf-promote-candidates is a skill published in the GitHub repository kerbelp/metatron (24 stars, last pushed 1mo ago), licensed MIT. It adds 56 tokens to every session and 1,259 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
shipping-a-pr
Use when finished work needs to ship as a pull request, or when the ask is about a PR — creating one, bringing it up to date, adding screenshots, watching its checks, or addressing its review comments. Not for reviewing a PR you are not shipping.
agentic-jujutsu
Quantum-resistant, self-learning version control for AI agents with ReasoningBank intelligence and multi-agent coordination.
agent-release-swarm
Agent skill for release-swarm - invoke with $agent-release-swarm.
agent-sync-coordinator
Agent skill for sync-coordinator - invoke with $agent-sync-coordinator.
agent-release-manager
Agent skill for release-manager - invoke with $agent-release-manager.
agent-repo-architect
Agent skill for repo-architect - invoke with $agent-repo-architect.