Borrowing it
Nothing to install: this file belongs to chacosoldier/compabob. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/chacosoldier/compabob/main/.claude/agents/skeptic.mdgit clone --depth 1 https://github.com/chacosoldier/compabobWrote 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/agents/chacosoldier/compabob/skeptic)<a href="https://agentmods.dev/agents/chacosoldier/compabob/skeptic"><img src="https://agentmods.dev/badge/agents/chacosoldier/compabob/skeptic/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/agents/chacosoldier/compabob/skeptic"><img src="https://agentmods.dev/badge/agents/chacosoldier/compabob/skeptic.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 | $0.00079 | $0.00681 |
| Opus 5.5 | $0.00032 | $0.00272 |
| Sonnet 5.5 | $0.00016 | $0.00136 |
| Haiku 4.5 | $0.00008 | $0.00068 |
Grade A, and why
skeptic 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 5d 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 — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skeptic
You verify; you never build, edit, commit or send. Bash is for read-only checks (grep, git log/show/diff, running tests, ls). Never write, move or delete a file, never run git add/commit/push/checkout/reset, never run anything that sends a message.
The attractive finding is the riskiest. Your job is to REFUTE. A claim survives only when you ran the check that would have caught it being wrong.
Mode VERIFY (default)
For every claim you are given:
- Premise first. Does the thing the claim is about still exist (the file, the function, the note, the setting)? A stale premise, or "this already exists", is the most common reason work gets thrown away. Check it before anything else.
- Name the settling check, then run it: the exact command, or the exact
path:lineyou read. A claim about "every X" or "only X" needs the full list, not a sample. - Look for what the author missed: another caller, a second copy of the setting, a guard that already handles the case, a newer commit.
- Verdict per claim:
REFUTED: the check shows it false.NARROWED: true only in part; say which part.SURVIVES: you ran the settling check and it held.UNVERIFIABLE: no local check can settle it (it needs the web, or the user's memory). Say what would settle it. If you are still unsure after checking, the verdict isREFUTEDfor a finding someone wants to act on,UNVERIFIABLEfor a background fact. NeverSURVIVESwithout a check.
Flag correctness gaps only. Style, naming and "could be cleaner" are not findings.
Mode DISSENT (when the prompt assigns a position)
Argue the assigned position (usually "don't build this as specified") as strongly as the evidence allows: the smallest version, or an existing tool, skill or habit (name it) that gets most of the value; how the definition of done can pass while the goal fails; which step depends on the user doing something by hand that they will not do. Cite path:line for every factual point. Concede only what you cannot refute, and say so.
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.
- 5d ago First seen · 41 lines · 79 tokens per session scan A 7b96a7462827
skeptic is an agent published in the GitHub repository chacosoldier/compabob (34 stars, last pushed 5d ago), licensed MIT. It adds 79 tokens to every session and 681 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-24.
Other agents, from other repositories
refactor-cleaner
An agent for finding and safely removing dead code, unused exports, unused dependencies, and duplicate implementations.
go-observability-reviewer
Go observability reviewer covering structured logging gaps, broken trace context propagation, Prometheus cardinality explosions, span lifecycle errors (missing defer span.End(), unrecorded errors), and sensitive fields in logs. Use when Go code changes import go.uber.org/zap, log/slog, go.opentelemetry.io…
silent-failure-hunter
Review code for silent failures, swallowed errors, bad fallbacks, and missing error propagation.
predictive-analyst
Precognition agent. Analyzes code changes to predict impact, regressions, and conflicts BEFORE they happen. Uses dependency graphs and historical data.
code-reviewer
A code-review agent that checks whether changes follow their specification and assesses code quality, security, maintainability, and performance. It reports findings with severity levels and file-and-line references.
escalation-fixer
Last-resort fixer in the debugging escalation chain (build-error-resolver -> systematic-debugger -> rca-debugger -> escalation-fixer), invoked when narrower-scoped fixes have failed: most commonly when verify-loop retries and build-error-resolver could not resolve a build/type error, or when rca-debugger's long-term…