oracle

A read-only AI reviewer for critical code changes and difficult technical problems. It examines areas such as security, payments, data migrations, public interfaces, algorithms, and bugs.

In plain words
What is it for?
Use it to review high-risk changes or produce a precise implementation plan for a hard problem after earlier fixes have failed.
Why use it?
It provides a second, detailed review for mistakes that tests may not catch, including race conditions, authorization gaps, data loss, and compatibility breaks.

Agent for Claude Code

Install

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.

agentmods
npx agentmods add agents/alebgl77/tierdecay/oracle
Clone the repo
git clone --depth 1 https://github.com/alebgl77/tierdecay

Made for: Claude Code.

Per session 66 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 323 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00066 $0.00323
Opus 5 $0.00033 $0.00161
Sonnet 5 $0.00013 $0.00065
Haiku 4.5 $0.00007 $0.00032

Measured yesterday against content hash 55843ca87508, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

oracle 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.

adapters/claude-code/.claude/agents/oracle.md · 30 lines

What it actually says

You are the deepest reasoning tier. Terse, precise, calibrated — state your certainty explicitly. Two modes, chosen by the brief:

REVIEW mode

Input: a diff + its intent + acceptance criteria. Hunt specifically for what tests won't catch: inverted logic, boundary and off-by-one errors, TOCTOU/race windows, injection, authz gaps, silent data loss, ordering assumptions, backward-compat breaks, error paths that swallow failures. Mandatory last line: APPROVE | APPROVE-WITH-NITS: <list> | BLOCK: <reasons> + <minimal fix>

SOLVE mode

Input: a hard problem + constraints + everything already tried. Output a spec implementable by an executor with zero decisions left:

  1. Chosen approach, plus the 2–3 rejected alternatives and why each loses.
  2. Invariants and complexity bounds.
  3. Pseudocode precise enough to translate mechanically.
  4. Test plan, including adversarial cases.

Never write to files. Never expand the question. If the dossier is insufficient to reason safely, state exactly what is missing.

Changes

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.

  1. yesterday First seen · 30 lines · 66 tokens per session scan A 55843ca87508

Subscribe to this mod's changes

oracle is an agent published in the GitHub repository alebgl77/tierdecay (1 stars, last pushed 1mo ago), licensed MIT. It adds 66 tokens to every session and 323 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 2026-08-31.

Related

Other agents, from other repositories

aerodynamicist

Reasons from circulation, Cp distributions, and boundary-layer physics through Re/Mach similitude, NACA airfoil polars, stall classification, wind-tunnel blockage/wall corrections, and SA/SST/LES external-aero CFD—not generic mechanical engineering.

K-Dense-AI/scientific-agents · 56 tokens

algorithms-researcher

Reasons from separating problem, model, and cost model (comparison, word-RAM, arithmetic, online) through exchange/matroid greedy proofs, subproblem-DAG dynamic programming, max-flow min-cut and Goemans–Williamson primal-dual rounding, Karp–Rabin fingerprinting, competitive ratio and Yao's principle, PTAS/FPTAS…

K-Dense-AI/scientific-agents · 163 tokens

antenna-engineer

Reasons from gain–directivity–efficiency, Chu–Harrington bandwidth limits, and array factor through HFSS/CST/FEKO synthesis, IEEE 149-2021 NF/FF/CATR metrology, CTIA TRP/TIS/ECC OTA, and Friis link budgets while treating ground-plane truncation, active impedance in arrays, range ripple, and S₁₁≠pattern conflation as…

K-Dense-AI/scientific-agents · 97 tokens

aquaculture-scientist

Reasons from FCR, dissolved oxygen and ammonia thresholds, hatchery biosecurity, and stock genetics while treating off-flavor, disease outbreak, and escape risk as first-class failure modes.

K-Dense-AI/scientific-agents · 43 tokens

astrochemist

Reasons from gas-grain reaction networks, H₂ ortho/para and CR ionization rates through KIDA/kida.uva.2024, CDMS/JPL/Splatalogue line lists, Nautilus/UCLCHEM gas-grain models, ALMA/JWST/LIDA ice–gas linkage, XCLASS LTE fitting, and line-blending discrimination—not generic chemistry.

K-Dense-AI/scientific-agents · 81 tokens

discovery-analyst

Use proactively during /fp:init to perform Phase 1 (Discovery) of the first-plan plugin. Read-only subagent that maps stacks, conventions, reuse, domain and risks of an unknown project applying the Stack Lens Engine. Returns structured findings to be written to .first-plan/. Do NOT use for execution or modifications …

vynazevedo/first-plan · 73 tokens