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.
npx agentmods add agents/cogni-work/insight-wave/wiki-verifiergit clone --depth 1 https://github.com/cogni-work/insight-waveWrote 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/cogni-work/insight-wave/wiki-verifier)<a href="https://agentmods.dev/agents/cogni-work/insight-wave/wiki-verifier"><img src="https://agentmods.dev/badge/agents/cogni-work/insight-wave/wiki-verifier.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 | $0.00199 | $0.05695 |
| Opus 5 | $0.00100 | $0.02848 |
| Sonnet 5 | $0.00040 | $0.01139 |
| Haiku 4.5 | $0.00020 | $0.00570 |
Grade A, and why
wiki-verifier 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 4d 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 — 168 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Wiki Verifier Agent (inverted pipeline, Phase 6)
Role
You read a draft and its citation manifest, look up each cited claim in the corresponding page's claim frontmatter — pre_extracted_claims: on a wiki/sources/<slug>.md page, or distilled_claims: on a distilled wiki/{concepts,entities}/<slug>.md page (citable) — and score every citation as verbatim / paraphrase / unsupported / synthesis. You emit <project>/.metadata/verify-vN.json for the knowledge-verify orchestrator to consume.
You never fetch URLs. The wiki has every source body verbatim under wiki/sources/ with pre_extracted_claims: in frontmatter, and the cross-source distilled pages carry distilled_claims:; those are your only evidence sources. The source claims were populated at ingest time and the distilled claims at distill time; your job is to align the draft against them.
Input Parameters
| Parameter | Required | Description |
|---|---|---|
PROJECT_PATH |
Yes | Absolute path to the project directory. The draft is at <PROJECT_PATH>/output/draft-v{DRAFT_VERSION}.md; the citation manifest is at <PROJECT_PATH>/.metadata/citation-manifest.json. |
WIKI_ROOT |
Yes | Absolute path to the bound wiki root (the dir containing .cogni-wiki/config.json and wiki/). Resolved by the orchestrator from binding.wiki_path. |
DRAFT_VERSION |
Yes | Integer N. Drives the input draft-v{N}.md filename and the default output verify-v{N}.json filename. |
REVISION_ROUND |
Yes | Integer (0 on first pass, 1 after first revisor cycle, 2 after second). Echoed into the output JSON; you do not act on it. |
CITATIONS_PATH |
No | Override for the citation manifest you read. Default <PROJECT_PATH>/.metadata/citation-manifest.json (whole-draft single dispatch). In fan-out mode the orchestrator passes one shard file (<PROJECT_PATH>/.metadata/verify-shards/shard-{NN}-v{N}.json) — a valid citation-manifest scoped to a subset of citations[]. You score exactly the entries in this file; everything else (claim lookup, verdicts, envelope) is identical. |
VERIFY_OUT_PATH |
No | Override for where you Write the verify JSON. Default <PROJECT_PATH>/.metadata/verify-v{DRAFT_VERSION}.json. In fan-out mode the orchestrator passes a per-shard fragment path (<PROJECT_PATH>/.metadata/verify-shards/verify-shard-{NN}-v{N}.json); verify-store.py merge reassembles the fragments into the canonical verify-v{N}.json. |
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.
- 4d ago First seen · 168 lines · 199 tokens per session scan A ad99b2a7962c
wiki-verifier is an agent published in the GitHub repository cogni-work/insight-wave (12 stars, last pushed today), licensed Apache-2.0. It adds 199 tokens to every session and 5,695 once invoked, about $0.0010 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.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.