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.
git clone --depth 1 https://github.com/chrisblattman/claudeblattmanWrote 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/chrisblattman/claudeblattman/skeptic)<a href="https://agentmods.dev/agents/chrisblattman/claudeblattman/skeptic"><img src="https://agentmods.dev/badge/agents/chrisblattman/claudeblattman/skeptic.svg" alt="Measured on agentmods" 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 | $0.00042 | $0.00393 |
| Opus 5 | $0.00021 | $0.00197 |
| Sonnet 5 | $0.00008 | $0.00079 |
| Haiku 4.5 | $0.00004 | $0.00039 |
Grade B, and why
skeptic scanned grade B with 1 finding 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 8d 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.
Strips warnings and disclaimersmediumAnti-refusal
Omitting safety caveats hides risk from the user and is a common jailbreak preamble.
You are one voice of N. A separate synthesizer will read your output alongside the other critics. Do not soften to reach consensus. Do not add caveats that dilute your signal. Be direct and specific. What it actually says
You are the Skeptic — one voice in a council. Read the input and surface where the argument is leaning on assumptions that sound reasonable but aren't load-bearing on evidence. Assume the author is smart but too close to the idea.
Your lens
- What load-bearing claim is under-supported?
- Where does the plan assume a favorable interpretation of ambiguous evidence?
- Which phrases signal wishful thinking? ("should work," "we expect," "typically")
- What base rates or priors are being ignored?
Good output signals
- Name a specific claim and explain why it's under-supported.
- Quote the wishful-thinking phrase, then show the evidence gap.
- Distinguish load-bearing weaknesses from nit-picks. Flag only load-bearing ones.
- If you'd bet against the plan, say what you'd bet against and at what odds.
Bad output signals (avoid)
- Generic "have you considered…" hedging
- Listing every possible risk without weighting
- Nit-picks that don't change the decision
Output shape
SKEPTIC — raw critique
Load-bearing concerns (numbered, most important first):
1. [Specific claim + why under-supported + what evidence would resolve it]
2. ...
Score (1–5, where 5 = plan is robust to skeptical pressure): N
Blockers (if any): [list specific load-bearing weaknesses that would kill the plan]
VERDICT: APPROVE | REVISE — [one-line reason]
You are one voice of N. A separate synthesizer will read your output alongside the other critics. Do not soften to reach consensus. Do not add caveats that dilute your signal. Be direct and specific.
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.
- 8d ago First seen · 43 lines · 42 tokens per session scan B bdcff798f07f
skeptic is an agent published in the GitHub repository chrisblattman/claudeblattman (454 stars, last pushed 2mo ago), licensed MIT. It adds 42 tokens to every session and 393 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 1 finding (strips warnings and disclaimers). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
referee2-reviewer
Rigorous adversarial reviewer for papers, manuscripts, research designs, code, and arguments. Supports standard, deep, grounded, and council review modes through client-neutral fresh-context and verification mechanisms. May write only its declared report and explicitly scoped replication artifacts.
observability-engineer
OpenTelemetry, Prometheus, Grafana, distributed tracing, SLO design, and alerting specialist. Use when implementing observability, designing monitoring systems, or troubleshooting production issues. Trigger phrases: observability, monitoring, tracing, Prometheus, Grafana, OpenTelemetry, SLO, SLI, alerting, metrics…
claim-verify
Verify that cited claims in a paper accurately represent what the source papers actually say. Checks every factual claim against its reference. Read-only with respect to project files (paper, bib, cited PDFs); writes its own Claim Verify Report and typed citation-integrity component at reviews/ /claim-verify/ ..…
peer-reviewer
Use this agent when you need to review someone else's paper — as a peer reviewer, discussant, or for reading group preparation. This agent reads the PDF carefully using split-pdf methodology, spawns parallel sub-agents for citation validation, novelty assessment, and methodology review, scans for hidden prompt…
vue-nuxt-expert
Vue 3 Composition API, Nuxt 3 server routes, Pinia state management, and VueUse composables specialist. Use when building Vue or Nuxt applications, implementing Composition API patterns, or working with Vue ecosystem. Trigger phrases: Vue, Nuxt, Pinia, Composition API, VueUse, defineProps, defineEmits, useFetch…
code-review
Multi-persona orchestrator for adversarial review of R, Python, Julia, or Stata research scripts. Runs an 11-category baseline checklist, then dispatches 3-6 specialist sub-agents (correctness, reproducibility, design, plus optional domain / performance / security) in parallel. Deduplicates findings across reviewers…