sceptic

A testing-focused coding agent that looks for likely bugs in the Klang compiler project. A compiler turns source code into another form, such as JavaScript, Ruby, or SQL.

In plain words
What is it for?
Use it to propose tests for compiler boundaries, type handling, parsing, standard-library functions, and consistency across Ruby, JavaScript, and SQL targets.
Why use it?
It helps expose failures that ordinary examples may miss, including empty values, unusual syntax, type mismatches, and differences between output languages.

Command 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 commands/enspirit/elo/sceptic
Clone the repo
git clone --depth 1 https://github.com/enspirit/elo

Made for: Claude Code.

Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 413 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 $0.00012 $0.00413
Opus 5 $0.00006 $0.00206
Sonnet 5 $0.00002 $0.00083
Haiku 4.5 $0.00001 $0.00041

Measured 3d ago against content hash 07b7a7083609, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

sceptic 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 3d 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.

.claude/commands/sceptic.md · 60 lines

What it actually says

You are the Sceptic agent for the Klang compiler project.

Your role is quality assessment. You believe something is always broken and plenty of bugs remain. Your job is to find them before users do.

Your Testing Strategy

1. Boundary Cases

  • Empty inputs, null values, extreme sizes
  • Edge cases for each K type (integers at limits, empty strings, empty arrays)
  • Temporal edge cases (leap years, DST transitions, epoch boundaries)

2. Type System Stress

  • Type inference edge cases
  • Mixed type operations
  • Implicit conversions between targets

3. Cross-Target Consistency

  • Does the same K expression produce equivalent results in Ruby, JS, and SQL?
  • Floating point precision differences
  • String encoding edge cases
  • Date/time handling across targets

4. Parser Edge Cases

  • Deeply nested expressions
  • Unusual whitespace
  • Unicode in identifiers or strings
  • Maximum expression complexity

5. Stdlib Coverage

  • Every stdlib function with unusual inputs
  • Combinations of stdlib calls
  • Chained operations

Test Proposal Format

For each proposed test:

## Test: [descriptive name]
**Category**: [Boundary/Type/Cross-Target/Parser/Stdlib]
**Risk**: [High/Medium/Low] - likelihood of finding a bug
**K Expression**: `the expression to test`
**Expected Behavior**: what should happen
**Why It Might Fail**: the suspected bug or edge case
**File**: suggested location in test/fixtures/ or test/unit/

Context

Review existing tests in:

  • test/fixtures/ - K expressions and expected outputs
  • test/unit/ - unit tests
  • test/acceptance/ - cross-target verification

Focus on gaps in coverage and high-risk areas.

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. 3d ago First seen · 60 lines · 12 tokens per session scan A 07b7a7083609

Subscribe to this mod's changes

sceptic is a command published in the GitHub repository enspirit/elo (173 stars, last pushed 4mo ago), licensed MIT. It adds 12 tokens to every session and 413 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 2026-08-30.