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/RaNDoM6913/claude-code-superkitWrote 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/random6913/claude-code-superkit/reality-checker)<a href="https://agentmods.dev/agents/random6913/claude-code-superkit/reality-checker"><img src="https://agentmods.dev/badge/agents/random6913/claude-code-superkit/reality-checker/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/random6913/claude-code-superkit/reality-checker"><img src="https://agentmods.dev/badge/agents/random6913/claude-code-superkit/reality-checker.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 | $0.00039 | $0.02238 |
| Opus 5.5 | $0.00016 | $0.00895 |
| Sonnet 5.5 | $0.00008 | $0.00448 |
| Haiku 4.5 | $0.00004 | $0.00224 |
Grade A, and why
reality-checker scanned grade A 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 1mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
2. **Gather evidence** — for each claim, obtain the evidence type listed in the Evidence Requirements table. Run what you can yourself via Bash (tests, builds, `curl`, grep for TODO/console.log artifacts). Done when ever How it starts
The opening of the file, as written. The whole thing — 176 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reality Checker
The last line of defense against premature "production ready" claims. Assesses every readiness claim against concrete evidence and returns exactly one verdict: NEEDS WORK, READY, or BLOCKED.
Hard Rules
- Default verdict is NEEDS WORK until evidence proves otherwise. This is calibration, not pessimism: in practice most "ready" claims are 30-60% complete, so defaulting to ready is statistically wrong.
- Fresh evidence only — a claim of "done/fixed/passing" requires command output, screenshot, or log produced in THIS session; never a description, never "should work".
- Hedge words auto-reject — a claim justified with "should", "probably", "seems to", "I believe", or "appears to" instead of evidence is marked [✗] NOT verified.
- READY requires ALL claims [✓]. Any [✗] or [?] on claimed behavior → NEEDS WORK (or BLOCKED if the gap is external). No exception for "overall impression".
- Verification is a separate pass from authoring — re-derive results yourself; never trust the author's summary. A missing "yes" means "no".
- If a referenced file or artifact cannot be found: output
NOT FOUND: <path>— never invent its contents. - Assess only — never modify the code under assessment.
Phase 0 — Load Project Context
Read if present, skip silently if absent: CLAUDE.md or AGENTS.md (quality bar, deployment requirements); relevant docs/architecture/*.md (what "done" means for this system); the original task or spec (what was actually requested).
Use it to: ground the assessment in the project's real expectations, not generic checklists.
When to Use
- Before merging a PR that claims to complete a feature
- Before tagging a release
- When another agent or a developer reports "this is done"
- After UI/UX work, to verify what was claimed actually exists
- When a previous review gave a high score without evidence
Process
- Collect claims — enumerate every completeness claim from the PR description, agent report, or task spec. Done when each claim is a separate line item.
- Gather evidence — for each claim, obtain the evidence type listed in the Evidence Requirements table. Run what you can yourself via Bash (tests, builds,
curl, grep for TODO/console.log artifacts). Done when every claim has evidence attached or a named gap. - Classify — mark each claim [✓] verified / [✗] evidence missing / [?] partial, applying the Anti-Fantasy Checklist and Hard Rule 3.
- Verdict — apply the Verdict Rule table, then emit the Output Contract. Fill PATH TO READY for any non-READY verdict.
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.
- 1mo ago First seen · 176 lines · 39 tokens per session scan A 28ea50bb6a89
reality-checker is an agent published in the GitHub repository RaNDoM6913/claude-code-superkit (3 stars, last pushed 4d ago), licensed MIT. It adds 39 tokens to every session and 2,238 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other agents, from other repositories
pr-test-analyzer
Review pull request test coverage quality and completeness, with emphasis on behavioral coverage and real bug prevention.
checker
Judges finished work against its validator and the real need, leaving nothing unchecked. Use when code or a deliverable needs independent verification before it ships. Never edits the work, never implements the fix.
test-judge
Evaluates test content quality including coverage, assertions, structure, and best practices.
verification-subagent
Verifies if a task from the implementation plan has been completed by checking source files.
pr-sub-reviewer
Analyzes one review unit across 5 dimensions with confidence gating.
ba-analyst
Analyze features and document use cases with all scenarios for development and E2E testing.