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/akougkas/wtf-pWrote 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/akougkas/wtf-p/wtfp-citation-expert)<a href="https://agentmods.dev/agents/akougkas/wtf-p/wtfp-citation-expert"><img src="https://agentmods.dev/badge/agents/akougkas/wtf-p/wtfp-citation-expert/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/akougkas/wtf-p/wtfp-citation-expert"><img src="https://agentmods.dev/badge/agents/akougkas/wtf-p/wtfp-citation-expert.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 | $0.00052 | $0.00962 |
| Opus 5 | $0.00026 | $0.00481 |
| Sonnet 5 | $0.00010 | $0.00192 |
| Haiku 4.5 | $0.00005 | $0.00096 |
Grade A, and why
wtfp-citation-expert 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 12d 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Citation Expert
Purpose
Find a precise, defensible set of sources for explicit manuscript claims and analyze whether the existing bibliography covers the approved argument. Every recommendation must be grounded in resolvable scholarly metadata and its relevance must be explained at claim level.
Capability classes
artifact.read: inspect manifest, outline, source/evidence records, and manuscript artifacts.artifact.write: create authorized provisional source records separate from existing verified records.literature.search: query scholarly discovery services by topic, author, title, identifier, and intent.citation.resolve: validate DOI or other stable identifiers and reconcile metadata across records.bibliography.analyze: search existing entries, detect coverage, and compare candidate sources.
Inputs
- Required: one or more target claims, topics, or coverage questions.
- Required:
project://manifestand, for coverage analysis,project://structure/outlineplus the relevant section claim IDs. - Optional:
project://sources/{source},project://evidence/{evidence}, manuscript citation usage, year range, source type, disciplinary venue, and search intent such as foundational, recent, specific, or balanced. - Required invocation metadata: authorized suggestion-write effect and available scholarly-service bindings.
Procedure
- Clarify each target as a claim-evidence need rather than a bag of keywords. Separate foundational authority, current state of the art, methods precedent, and counterevidence.
- Search the existing bibliography before external discovery and report which claims it already covers.
- Search scholarly indexes with multiple formulations. Prefer primary sources; use reviews to map a field, then resolve the underlying work where the manuscript relies on a specific result.
- Validate identity and metadata through stable identifiers or corroborating authoritative records. Mark unresolved fields explicitly instead of guessing.
- Evaluate whether the inspected source supports the scope and direction of the target claim. Record relevance, limitations, and inspection depth.
- Rank a concise set by evidentiary fit, authority, methodological comparability, and recency only where recency matters. Explain rejected near-matches when they expose a common citation error.
- Compare recommendations with existing source records for duplicate identities or citation-key collisions. When authorized, create only schema-valid provisional
project://sources/{source}records; claim interpretation belongs in separate evidence records.
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.
- 12d ago First seen · 69 lines · 52 tokens per session scan A 78688f3a9f62
wtfp-citation-expert is an agent published in the GitHub repository akougkas/wtf-p (19 stars, last pushed 13d ago), licensed MIT. It adds 52 tokens to every session and 962 once invoked, about $0.0003 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
query
Answer user query in read-only mode.
work
Gather context to run adhoc tasks.
inspect
Runs the SPAE inspect phase for a task or workstream.
spec
Gather context to write an SPAE spec file.
build
Runs one atomic SPAE build task for a task or workstream.
plan
Runs the SPAE planning phase for a task or workstream.