self-eval-bias

A review practice that checks whether an agent is judging its own recent work fairly. It looks for cases where the same reasoning that created a plan, code change, or report also excuses its problems.

In plain words
What is it for?
Use it before marking work complete, accepting a positive review, closing an issue, or passing results between agents.
Why use it?
It helps avoid approving work just because it sounds convincing in the context that produced it.

Skill for Claude CodeCodex

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 skills/archive228/loopkit/self-eval-bias
Any agent
npx skills add Archive228/loopkit --skill self-eval-bias
Clone the repo
git clone --depth 1 https://github.com/Archive228/loopkit

Made for: Claude Code, Codex.

Per session 37 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 900 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 $0.00037 $0.00900
Opus 5 $0.00018 $0.00450
Sonnet 5 $0.00007 $0.00180
Haiku 4.5 $0.00004 $0.00090

Measured 2d ago against content hash 7d03635ea2f3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

self-eval-bias 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 2d 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.

skills/self-eval-bias/SKILL.md · 46 lines

How it starts

The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Self-Eval Bias

An agent that just produced a plan, a diff, or a report cannot fairly grade it in the same context. The reasoning that justified writing it is still loaded — every doubt was already resolved in favor of shipping. Asked to review, the same context reliably returns "looks good, ship it." This is not review. It is rationalization wearing a review's uniform.

The pattern shows up hardest in planner/generator/evaluator architectures where the evaluator drifts toward leniency over long runs — the prompts it reads fill up with the generator's reasoning, and skepticism erodes. (See Prithvi's March 2026 post on the three-agent harness: https://blog.anthropic.com/three-agent-harness-march-2026.)

When to apply

  • You just wrote code, a plan, or a claim, and the next step is "confirm it's correct".
  • A reviewer verdict comes back positive with no cited line numbers, no failing case explored, no counter-example attempted.
  • You're about to mark a feature passes: true, close an issue, or hand off to the next session.
  • The evaluator persona in a multi-agent loop has agreed with the last N generator outputs in a row.

Procedure

  1. Notice the same-context tell. If the review verdict lands in under three sentences and contains "looks correct", "this should work", or "no issues found" without a cited artifact — treat the verdict as unwritten.
  2. Force a fresh persona. Drop the generation context. Open a new subagent, or at minimum re-prompt with only the artifact (diff, plan, output) and the acceptance criteria — no reasoning trail, no self-justification.
  3. Demand concrete evidence, not verdicts. The reviewer must cite: the file:line it inspected, the input it ran, the observed output, and the criterion it matched against. "LGTM" without these is a null review — discard it.
  4. Adversarially probe. Ask the reviewer for the strongest case where the artifact fails. If it can't produce one, the review didn't happen — the reviewer just agreed.
  5. Run the artifact. For code, exercise it end-to-end (see [[broken-window-check]]). For a plan, walk the first two steps concretely. Same-context confidence collapses fast against a runtime.
  6. Rotate the reviewer periodically. In long multi-agent loops, re-prompt the evaluator from scratch every ~5 sprints — leniency drift compounds silently.

Read the full file on GitHub · 46 lines

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. 2d ago First seen · 46 lines · 37 tokens per session scan A 7d03635ea2f3

Subscribe to this mod's changes

self-eval-bias is a skill published in the GitHub repository Archive228/loopkit (753 stars, last pushed 1mo ago), licensed MIT. It adds 37 tokens to every session and 900 once invoked, about $0.0002 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 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens