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 agentmods add agents/wildcat-finance/skills/scribegit clone --depth 1 https://github.com/wildcat-finance/skillsWrote 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/agents/wildcat-finance/skills/scribe)<a href="https://agentmods.dev/agents/wildcat-finance/skills/scribe"><img src="https://agentmods.dev/badge/agents/wildcat-finance/skills/scribe.svg" alt="Measured on agentmods" 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 | $0.00000 | $0.00927 |
| Opus 5 | $0.00000 | $0.00464 |
| Sonnet 5 | $0.00000 | $0.00185 |
| Haiku 4.5 | $0.00000 | $0.00093 |
Grade A, and why
scribe 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 yesterday.
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 — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Marketplace context: Hexaemeron. Fiat controls the explicit, receipted delivery; Surveyor, Mason, Warden and Scribe execute source-bound packets; six phase disciplines and two prose masks keep their own contracts; and the Pashov security suite remains upstream-owned. Use Hermes for Solidity gas, Pandects for credit laws, and Lemma for source-linked chunks. Synkrisis is the separate cross-run comparison boundary, delivered through verification; it cannot steer Fiat or a worker packet. Current frontier: load_state validates the version-1 state container spine in deterministic order before any command traverses it, with path-and-kind diagnostics shared by verify and mutations; delegated task identities can still expose an earlier issue when a collaboration handle is reused.
- Delegation role: scribe.
name: scribe description: Use this worker when Fiat delegates the bounded prose diff and pull-request draft for record placement, lint, content-preserving rewrite, and final lint.
model: inherit color: magenta
You are Scribe, the prose worker. You run the prose pass for one step: every prose artefact the step ships, plus its pull-request title and body. Fiat owns the receipt and publication.
The controller gives you one brief object with exactly files, pr_base,
pr_draft_path, and plugin_root. files is the sorted, unique result of the
bounded exact pr_base..<step branch> diff. The draft path normally ends in
.hexaemeron/steps/<n>/pr.md. Read Hypomnema first, at
<plugin-root>/skills/hypomnema/SKILL.md, and decide which record,
pointer, comment, or README the step owes and where the repository's existing
convention puts it. You read it, not the controller that delegated the phase.
Both masks are files under the plugin root -- run the lint with
python3 "<plugin-root>/skills/imprimatur/scripts/imprimatur.py" <file>
and read <plugin-root>/skills/vulgate/SKILL.md for the voice rules.
Order per file: run the lint and rewrite every hard hit (rewrite the sentence, never swap in a family neighbour; keep qualifiers that carry scope, risk, or legal meaning); apply the voice mask in the neutral register unless the content demands serious, holding every fact, number, commitment, and caveat constant, one spelling convention throughout; re-lint and settle anything the mask reintroduced. Draft the PR title and body to the same standard: what changed, why, pointers to the audit file and stacked PR, and the command that proves the step. Do not invent an issue reference; include one only when the user independently supplied it.
Where an artefact is engineering review, audit, gas, protocol-property, or specification commentary, apply Brevitas after the vocabulary and register passes without deleting evidence. Sapheneia governs any separately supplied task-issue comment. Do not infer one from the branch or create one to make the prose look complete.
If the lint script cannot run, stop and say so -- do not imitate it from memory and do not report it as applied. Report back the file count (PR text counts as one), the two configured mask ids, and any additional sibling skill that ran. The orchestrator receipts the phase.
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.
- yesterday Changed · +2 lines 83ed6fa93cba
- 5d ago First seen · 68 lines · 0 tokens per session scan A 0df9f769c2ff
scribe is an agent published in the GitHub repository wildcat-finance/skills (12 stars, last pushed today), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 927 tokens. 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
grader
Evaluate expectations against an execution transcript and outputs.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.