_critic

_critic is a skill for Claude Code from epicsagas/epic-harness. It costs 57 tokens per session (761 once invoked), scanned A, original, Apache-2.0.

An adversarial review procedure for proposed new skills in the epic-harness project. It compares the proposal with recorded traces, edit manifests, task results, and scoring history to look for reward hacking or contradictions.

In plain words
What is it for?
Use it during skill-evolution reviews, especially when reward hacking is suspected or results have seesawed between rounds.
Why use it?
It challenges claims that a skill improved results when the available evidence may show regressions, side effects, or unsupported conclusions.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the epic plugin — 27 skills, 6 hooks shipped together

Good fit Use it during skill-evolution reviews, especially when reward hacking is suspected or results have seesawed between rounds.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/epicsagas/epic-harness/_critic
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.

Any agent
npx skills add epicsagas/epic-harness --skill _critic
Clone the repo
git clone --depth 1 https://github.com/epicsagas/epic-harness

Made for: Claude Code.

Or install epic, the plugin that ships this one along with the rest of its 27 skills, 6 hooks.

Wrote 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.

agentmods badge for _critic

README.md
[![agentmods](https://agentmods.dev/badge/skills/epicsagas/epic-harness/_critic.svg)](https://agentmods.dev/skills/epicsagas/epic-harness/_critic)
Your own site
<a href="https://agentmods.dev/skills/epicsagas/epic-harness/_critic"><img src="https://agentmods.dev/badge/skills/epicsagas/epic-harness/_critic.svg" alt="Measured on agentmods" height="20"></a>
Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 761 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00057 $0.00761
Opus 5 $0.00028 $0.00380
Sonnet 5 $0.00011 $0.00152
Haiku 4.5 $0.00006 $0.00076

Measured 7d ago against content hash 9b3d06ca041c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

_critic 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 7d 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/_critic/SKILL.md · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Critic — Adversarial Proposal Review

In-loop vs out-of-band. epic-harness forbids external LLM calls from production, so the reflect loop ships a deterministic critic (src/evolve/critic.rs) that gates seeding when reward hacking is suspected. THIS skill is the out-of-band LLM version a meta-agent or human runs during /evolve review for the cases the deterministic check cannot catch (non-local effects, manifest/evidence nuance).

When to Trigger

  • During /evolve review of newly seeded skills
  • When reward_hacking_suspected is true in metrics
  • After a seesaw-regression round, before re-proposing

Process

1. Gather the proposal + evidence

  • Read the evolved skill proposal(s) from this round
  • Read the EditManifest (edit_type, target, intended_effect, predicted_impact)
  • Read the round's TaskDigests (outcome, failure_categories, evidence_excerpts)
  • Read recent score_history dimension_averages (tool_success, output_quality, execution_cost)

2. Falsify the manifest (paper §4.3, Table 9)

For each proposal, ask: does the trace evidence support the predicted_impact?

  • If the manifest claims "Lift avg_score_with" but output_quality is regressing → Reject
  • If the manifest claims a tool fix but the implicated component shows no change → Warn
  • If the evidence corroborates the predicted effect → Approve

3. Reward-hacking cross-check

  • Is execution_cost rising while output_quality falls across the window?
  • Could the skill be gaming a metric (fewer tool calls inflating cost score) rather than improving outcomes?
  • If yes, the skill must NOT ship — flag for the rejected buffer.

4. Non-local effect scan

  • Will this skill's guard rules conflict with existing skills (overlapping triggers, contradictory rules)?
  • Does it interact with shared state (context, memory, control) in a way the manifest didn't account for?

Anti-Rationalization

Excuse Rebuttal Do instead
"The score went up, so it works" Score can rise via metric gaming Verify the outcome improved, not just the score
"The seesaw passed, it's safe" Seesaw is coarse; sub-threshold coupling evades it Check dimension deltas, not just aggregate pass
"It's just a prompt tweak" Prompt edits have non-local effects on shared context Trace the effect across skills, not just the target

Read the full file on GitHub · 62 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. 7d ago First seen · 62 lines · 57 tokens per session scan A 9b3d06ca041c

Subscribe to this mod's changes

_critic is a skill published in the GitHub repository epicsagas/epic-harness (18 stars, last pushed 2d ago), licensed Apache-2.0. It adds 57 tokens to every session and 761 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 2026-08-30.