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 skills/timsimpsonjr/magpie/entity-graphnpx skills add TimSimpsonJr/magpie --skill entity-graphgit clone --depth 1 https://github.com/TimSimpsonJr/magpieWrote 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/skills/timsimpsonjr/magpie/entity-graph)<a href="https://agentmods.dev/skills/timsimpsonjr/magpie/entity-graph"><img src="https://agentmods.dev/badge/skills/timsimpsonjr/magpie/entity-graph.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 | $0.00069 | $0.03679 |
| Opus 5 | $0.00034 | $0.01840 |
| Sonnet 5 | $0.00014 | $0.00736 |
| Haiku 4.5 | $0.00007 | $0.00368 |
Grade A, and why
entity-graph 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 — 298 lines — stays where its author put it; the contents beside it link to each section on GitHub.
entity-graph
Take the per-document reviewed FtM bundles Phase 12 produced, resolve which entities ACROSS documents are the same real-world person or organization, let a human decide the uncertain matches, and write the resolved network to a Neo4j graph scoped to one investigation. This is the FIRST heavy-infra (Docker) phase.
This is a Layer-2, OPERATOR-tier, Docker-gated workflow. The journalist onramp
(JOURNALIST_START, the doctor skill's Track-A capabilities) stays Docker-free
and is NOT touched by this skill. Do not pull this flow into any journalist
surface.
The differentiator is the MANDATORY HUMAN REVIEW GATE, not the matcher. The matcher (nomenklatura LogicV2, a model-free heuristic) is imperfect on FOIA names; what makes the resolved graph trustworthy is that a person actively decides every uncertain pair and every auto-merge is logged and reversible. NEVER run this flow autonomously past the review gate.
13a (this skill) does resolution + the Neo4j write. Watchlist / sanctions / PEP
cross-reference and own-corpus cross-ref (yente + OpenSearch + the yente-mcp
server) are the SEPARATE entity-crossref skill (Phase 13b) -- run it AFTER this
one, on the resolved snapshot. Do not attempt cross-ref here.
What you need before you start
- Phase-12 FtM bundles for the corpus documents (the
<name>.entities.ftm.json- sibling
<name>.provenance.jsonl+<name>.manifest.jsontriples thatentity_ftmize.write_bundleproduced). Resolution runs ACROSS the whole investigation corpus at once -- pass ALL the bundles together, not one per document. That cross-corpus pass is what lets a homonym in two different documents surface as one review candidate.
- sibling
- The Phase-6 source page text for the same documents, reachable through a
snippet_resolvercallable (see STEP 2) -- the review packet hydrates each candidate's source snippet from it at render time. - Docker + Neo4j running (the PRECONDITION below).
- A per-investigation scratch directory (gitignored). The resolver SQLite DB, the
candidate snapshot, the auto-merge log, and
run.jsonall live here. It is per-investigation on purpose (global resolution would contaminate unrelated cases). If a run crashes mid-way, just DISCARD the scratch DB and re-run -- it is disposable.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 298 lines · 69 tokens per session scan A ceee69b97c40
entity-graph is a skill published in the GitHub repository TimSimpsonJr/magpie (2 stars, last pushed 2mo ago), licensed MIT. It adds 69 tokens to every session and 3,679 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 2026-08-31.
Other skills, from other repositories
pdf-table-extractor-brief
Produces a structured extraction plan and clean spreadsheet template for pulling tabular data out of a PDF document — identifying the table structure, defining column headers, flagging extraction pitfalls, and providing a ready-to-use template that ensures the data lands in a consistent, analysable format.
foia-request-writer
Drafts legally complete public records requests (federal FOIA and all 50 state laws), administrative appeals, and redaction challenge strategies for U.S. government records.
osint-tool-catalog
Produces a categorised catalog of open-source intelligence tools relevant to a journalist's investigation, with practical guidance on what each tool does, when to use it, and what its limitations are.
social-media-intelligence
Produces a structured open-source intelligence brief on a social media account or set of accounts, covering account authenticity analysis, narrative tracking, and coordination-detection patterns to support investigative reporting.
data-cleaning-brief
Writes clear, step-by-step instructions for cleaning a messy or inconsistent dataset — specifying exactly what needs to be standardised, corrected, or removed to make the data ready for analysis and publication.
data-outlier-finder
Identifies unusual values, unexpected patterns, and potential stories hidden in a dataset by systematically checking for statistical outliers and contextual anomalies.