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 panjose/Co-Scientist --skill ranking-elo-updategit clone --depth 1 https://github.com/panjose/Co-ScientistWrote 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/panjose/co-scientist/ranking-elo-update)<a href="https://agentmods.dev/skills/panjose/co-scientist/ranking-elo-update"><img src="https://agentmods.dev/badge/skills/panjose/co-scientist/ranking-elo-update/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/panjose/co-scientist/ranking-elo-update"><img src="https://agentmods.dev/badge/skills/panjose/co-scientist/ranking-elo-update.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 | $0.00026 | $0.00903 |
| Opus 5 | $0.00013 | $0.00451 |
| Sonnet 5 | $0.00005 | $0.00181 |
| Haiku 4.5 | $0.00003 | $0.00090 |
Grade A, and why
ranking-elo-update 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 10d 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 — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ranking-elo-update
Goal:
- Apply deterministic Elo updates for one completed tournament batch, persist the touched hypothesis artifacts, and write the ranking update receipt.
Inputs:
- completed tournament
matches - canonical
matchups strategyrun_dir- optional
k_factor - optional
top_k_limit
Outputs:
- updated in-memory hypothesis ratings and match references
- updated
hypotheses/<id>/HYPOTHESIS.jsonartifacts for all touched hypotheses state/ranking_update_receipts/<receipt_id>.jsonRankingUpdateReceiptContract
Context Loading:
- Open
skills/shared-references/schema-index.md. - Read
packages/agent_contracts/ranking.pyand confirm the exactTournamentMatchContractplusHypothesisMatchupContractshapes. - Read
packages/agent_contracts/hypothesis.pyand confirm the exactHypothesisContractshape before writing any touchedHYPOTHESIS.jsonartifact. - Use only tournament artifacts whose winners have already been decided and serialized.
Execution Contract:
- This skill is deterministic and must not call an LLM.
- Use
from tools import apply_and_persist_elo_updatesas the stable invocation surface. - The exported closeout helper is implemented in
packages/run_artifacts/ranking_writeback.py. - The helper signature is
apply_and_persist_elo_updates(run_dir, matches, matchups, strategy, k_factor=..., top_k_limit=...) -> RankingUpdateReceiptContract. - The closeout helper internally calls the canonical Elo helper, persists touched hypotheses, and writes the ranking update receipt.
- The closeout helper is idempotent for the same stable receipt ID: if the receipt already exists, it returns the persisted receipt instead of reapplying Elo deltas or appending duplicate match refs.
- Accepted
strategyvalues areplacement_tournamentandranked_tournamentonly. - The closeout helper updates canonical hypothesis bundle fields for touched hypotheses, but it must not rewrite standalone
REVIEW/*.jsonstage artifacts. - Match refs are not a ranking update receipt. Completion requires the persisted
RankingUpdateReceiptContract.
Execution Steps:
- Open
skills/shared-references/schema-index.md, then readpackages/agent_contracts/ranking.pyandpackages/agent_contracts/hypothesis.pybefore writing any touchedHYPOTHESIS.json. - Load the completed tournament matches as
TournamentMatchContractpayloads and the paired matchup payloads asHypothesisMatchupContract. - Confirm every match already has a decided
winner_idand that the requested strategy is valid. - Call
tools.apply_and_persist_elo_updates(run_dir, matches, matchups, strategy, k_factor=..., top_k_limit=...). - Confirm the returned
RankingUpdateReceiptContractnames every completed match ID and touched hypothesis ID. - Validate the touched hypothesis artifacts and
state/ranking_update_receipts/<receipt_id>.jsonbefore declaring completion.
Artifact Rules:
- This skill updates ratings only after the tournament winner artifacts are already finalized.
- Placement rounds update
placement_match_idsfor the challenger hypothesis andranked_match_idsfor the defender. - Ranked rounds update
ranked_match_idsfor both participating hypotheses. - Match ref arrays must remain duplicate-free. If a completed match ID is already recorded by both touched hypotheses for the same strategy, do not apply the Elo delta again.
- Do not invent extra ranking heuristics outside the helper; the helper is the canonical Elo update rule.
- Do not replace canonical ranking closeout with ad hoc file edits; use
tools.apply_and_persist_elo_updates(...)after every Elo batch. - Review-stage synchronization is a separate concern owned by
tools.sync_hypothesis_review(...); Elo writeback must not clobber those standalone review artifacts. - Do not hand-write ranking update receipts. The closeout helper owns receipt creation.
Completion Rule:
- This skill is complete only when the closeout helper has been applied once to the finished batch, every touched
hypotheses/<id>/HYPOTHESIS.jsonartifact has been rewritten in canonical contract form, and the corresponding ranking update receipt exists.
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.
- 10d ago First seen · 70 lines · 26 tokens per session scan A e0df1e183fef
ranking-elo-update is a skill published in the GitHub repository panjose/Co-Scientist (5 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 26 tokens to every session and 903 once invoked, about $0.0001 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
figure-style
Publication-grade correctness and legibility rules for final-deliverable scientific figures, not exploratory plots. Use for a figure that will ship in a report, paper, export, or kept artifact. Covers data fidelity, label economy, color threading, chart choice, layout, and render-then-verify QA without imposing a…
remote-compute-ssh
Evaluate and use SSH Remote Compute before choosing where to run GPU, high-memory, parallel, batch, model-inference, bioinformatics, or other long-running scientific work; supports short remote commands and asynchronous jobs with automatic harvest and analysis.
scvi-tools
Probabilistic single-cell RNA-seq with scvi-tools — scVI for a batch-corrected latent space, scANVI for semi-supervised label transfer, and Bayesian differential expression. Reach for this skill to integrate scRNA-seq batches, embed cells for clustering, transfer annotations from a reference onto a query, or score…
literature-review
Find, verify, and synthesize scientific literature — from "what's the seminal paper for X" through full multi-source reviews. Covers grounding claims in real retrieved sources, avoiding fabricated citations, handling retractions, and calibrating confidence to evidence strength.
openfold3
Structure prediction using OpenFold3, an open-weights PyTorch reproduction of AlphaFold3 from the AlQuraishi Lab. Use this skill when predicting protein/nucleic-acid/ligand complex structures with an Apache-2.0-licensed AF3 reimplementation.
self-awareness
Inspect Open Science's JavaScript control REPL, discover managed Project files, Sessions, and Agent Frames, and safely feature-gate host. calls with host.capabilities(). Use when an Agent needs to discover available host APIs, locate an Artifact or Upload Version, diagnose a Session, or read a Frame transcript in the…