Borrowing it
Nothing to install: this file belongs to Lengros/bennu. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Lengros/bennu/main/.claude/skills/evidence-ledger/SKILL.mdgit clone --depth 1 https://github.com/Lengros/bennuWrote 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/lengros/bennu/evidence-ledger)<a href="https://agentmods.dev/skills/lengros/bennu/evidence-ledger"><img src="https://agentmods.dev/badge/skills/lengros/bennu/evidence-ledger/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/lengros/bennu/evidence-ledger"><img src="https://agentmods.dev/badge/skills/lengros/bennu/evidence-ledger.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 | $0.00169 | $0.02523 |
| Opus 5 | $0.00084 | $0.01262 |
| Sonnet 5 | $0.00034 | $0.00505 |
| Haiku 4.5 | $0.00017 | $0.00252 |
Grade A, and why
evidence-ledger 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 9d 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 — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/evidence-ledger — living portfolio of risky assumptions
Maintains one artifact that tracks every load-bearing assumption a product bet depends on, in test order, with a threshold pre-registered before the data is seen and a forced decision after. It does not run the experiments. It keeps you honest about what each experiment can prove and stops you moving the goalposts once results arrive.
The engine, repeated at every stage: a believable surface is now free to produce, so the only thing that separates a real bet from a pretty one is a falsifiable threshold set in advance and a signal with something at stake. Curiosity is not commitment; an AI summary is not a finding; a threshold chosen after the data is rationalization.
Argument (optional): the bet/outcome, or a path to an existing ledger to update. If absent, use the bet under discussion in context.
Relationship to /reframe: /reframe regresses one request to a falsifiable outcome (the
INTENT stage). /evidence-ledger sits one level up — it holds the set of assumptions behind
a bet and sequences the experiments. A single assumption's experiment can itself be run
through /reframe. Use reframe for "what does this task have to achieve"; use
evidence-ledger for "what do we actually know about this bet, and what do we test next."
How to run
Work the seven stages in order. Append to the ledger artifact (§Schema); never rewrite a pre-registered threshold after a signal lands — supersede the row and keep the original.
-
Frame the bet — one line: the outcome the whole ledger serves (a result in the world, not a feature). May be lifted from a prior
/reframeoutcome. -
List load-bearing assumptions — everything that must be true for the bet to pay off. State each so it can be wrong ("tutors will write up every lesson the day it happens"), not as a goal ("great onboarding"). Surface the assumptions that would kill the bet if false, not the convenient ones you already believe.
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.
- 9d ago First seen · 152 lines · 0 tokens per session scan A 526b0c337e8c
evidence-ledger is a skill published in the GitHub repository Lengros/bennu (2 stars, last pushed 2mo ago), licensed MIT. It adds 169 tokens to every session and 2,523 once invoked, about $0.0008 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
dcf-model
Build discounted cash flow valuation workbooks in Excel.
hermes-agent-skill-authoring
Author in-repo SKILL.md files: frontmatter and structure.
systematic-debugging
4-phase root cause debugging: understand bugs before fixing.
openhands
Delegate coding to OpenHands CLI (model-agnostic, LiteLLM).
baoyu-article-illustrator
Article illustrations: type × style × palette consistency.
flash-attention
Speed up long-sequence transformer training and inference.