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 skills/yaniv-golan/proof-engine/proof-engine-wikinpx skills add yaniv-golan/proof-engine --skill proof-engine-wikigit clone --depth 1 https://github.com/yaniv-golan/proof-engineWrote 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/yaniv-golan/proof-engine/proof-engine-wiki)<a href="https://agentmods.dev/skills/yaniv-golan/proof-engine/proof-engine-wiki"><img src="https://agentmods.dev/badge/skills/yaniv-golan/proof-engine/proof-engine-wiki.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 | $0.00078 | $0.00388 |
| Opus 5 | $0.00039 | $0.00194 |
| Sonnet 5 | $0.00016 | $0.00078 |
| Haiku 4.5 | $0.00008 | $0.00039 |
Grade A, and why
proof-engine-wiki 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 4d 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.
What it actually says
proof-engine-wiki
Hooks Proof Engine verification into LLM-wiki ingest and lint operations.
When to use
Invoke this skill when:
- A new source has been added to a wiki and its draft pages contain
{{prove:}}markers. - A periodic lint pass is running — re-verify citations, flag contradictions.
- A user asks "is this claim actually proven?" while editing a wiki page.
When NOT to use
- To write proofs from scratch — that is the
proof-engineskill (the parent). - To extract claims from unmarked prose — this skill only processes explicit
{{prove:}}markers. Auto-extraction is out of scope for v0.1.
Workflow
- Read the wiki page or directory.
- Run
proof-engine-wiki ingest <path>(orlint, as appropriate). - Review the output plan; commit the rewritten pages.
Full runbooks in references/ingest-runbook.md and references/lint-runbook.md.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 4d ago First seen · 42 lines · 78 tokens per session scan A 94e237a736a0
proof-engine-wiki is a skill published in the GitHub repository yaniv-golan/proof-engine (8 stars, last pushed 3mo ago), licensed MIT. It adds 78 tokens to every session and 388 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
magic-compact
Compact this Claude Code session.
fulcra-media
Import media history (Watched / Listened / Read) into a Fulcra account via the fulcra-media CLI. Use when the user wants to track Netflix, Trakt, Spotify, Last.fm, Apple Podcasts, Letterboxd, Goodreads, YouTube, Plex/Jellyfin, or any other media source listed in fulcra-media import --help.
sealed-secrets
Use when an agent or role needs credentials on a machine that does not have them — a fresh container, a rebuilt host, a successor session — or when sealing, rotating, or verifying secrets kept in a shared object store.
fulcra-csv
Import any CSV stream into a Fulcra account as annotations — body weight, mood scores, expenses, sleep, media plays, anything timestamped. Use when the user wants to ingest a CSV they got from somewhere (a wearable, a Pipedream/IFTTT workflow, a hand-rolled spreadsheet) into Fulcra.
fulcra-agent-atc
Model & cap routing (ATC) for a subscription fleet: declare your accounts once, then a capability-ranked router (coord-engine route --needs …) picks the cheapest model that can do the job across every subscription, a cross-account cap ledger keeps traffic inside its windows, and outcome logging demotes a model that…
fulcra-agent-cloud-coordinator
Configure a Claude Code cloud session as a persistent, container-reset-proof coordinator that loops on schedules and keeps ALL state on the Fulcra bus — the pattern a live fleet coordinator has run in production.