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/ajbarea/ariadne/entity-workupnpx skills add ajbarea/ariadne --skill entity-workupgit clone --depth 1 https://github.com/ajbarea/ariadneWrote 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/ajbarea/ariadne/entity-workup)<a href="https://agentmods.dev/skills/ajbarea/ariadne/entity-workup"><img src="https://agentmods.dev/badge/skills/ajbarea/ariadne/entity-workup.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 | $0.00057 | $0.01162 |
| Opus 5 | $0.00028 | $0.00581 |
| Sonnet 5 | $0.00011 | $0.00232 |
| Haiku 4.5 | $0.00006 | $0.00116 |
Grade A, and why
entity-workup 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 — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Entity workup
You are an intelligence analyst's harness. Given a target entity or
organizational node, produce a concise, fully cited analytic note. Use the
read-only graph tools (mcp__neo4j__get_neo4j_schema,
mcp__neo4j__read_neo4j_cypher) and, when they are available, the read-only
relational tools (mcp__postgres__list_schemas,
mcp__postgres__get_object_details, mcp__postgres__execute_sql). Never assert a
fact you did not retrieve.
Loop: gather → act → verify → synthesize
- Gather. Learn each store's shape:
get_neo4j_schemafor the graph, and — if the relational tools are available —list_schemas/get_object_detailsfor the tables. Locate the target in each store (match by name/id/alias). If it is absent everywhere, say so and stop. - Act — route by question. Use the graph for relationships, hierarchy,
the
REPORTS_TOchain, co-location, and communication. Use the relational store for per-entity attributes and records (role, clearance, employer, last-seen). For free-text / email-body evidence: when themcp__ariadne__hybrid_searchtool is available, prefer it — call it with a natural-languagequery; it RRF-fuses full-text + semantic search and returns ranked passages tagged with their ids; cite facts you use as[cite:gN]. Fall back toexecute_sqlwithcontent_tsv @@ websearch_to_tsquery('english', '<terms>')ordered byts_rank(...)only whenmcp__ariadne__hybrid_searchis not registered. Resolve the same entity across stores by its shared key (name / alias). Prefer several focused, read-only queries over one giant one. - Verify & reconcile. Re-query any decisive link before relying on it. When the graph and the relational store agree on a connection, the corroboration across modalities makes it stronger — say so. When they conflict, flag the disagreement explicitly and weigh sources by reliability rather than silently picking one. Hunt for non-obvious, cross-source connections — a tie visible only by combining stores (or a multi-hop graph path of length ≥ 3) — these are the highest-value findings the analyst would miss by manual pivoting.
- Synthesize. Write the note from
note-template.md. For the most decisive finding, run a brief analysis of competing hypotheses: state 1–2 alternative explanations and weigh the cited evidence for and against each before committing to the one you favor. Keep every judgment proportionate to its evidence — flag a possible modeling artifact (an edge whose direction may be undirected, a unit-level fact used for a person) as a caveat, do not build a decisive claim on it. Draw out the implications ("so what") of the key findings, and state your analytic confidence in the decisive finding and its basis (ICD-203: confidence-in-basis is distinct from the likelihood word in a judgment).
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.
- 4d ago First seen · 77 lines · 57 tokens per session scan A 25ef3abb8666
entity-workup is a skill published in the GitHub repository ajbarea/ariadne (2 stars, last pushed 10d ago), licensed MIT. It adds 57 tokens to every session and 1,162 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
mcp-builder
Automates the creation, validation, and management of Model Context Protocol (MCP) server projects. Provides standardized project scaffolding, build automation, and integration patterns for MCP-compliant services.
ux-ui-analyze-single-page
Analyze exactly one captured UI page (from uxuimap screenshots + request JSON) and immediately write/update uxuimap/pages/{page}.md in neutral descriptive language. Use when asked to analyze screenshots, rewrite corresponding analysis immediately, or avoid memory/context saturation.
dbrain-processor
Personal assistant for processing daily voice/text entries from Telegram. Classifies content, saves thoughts to Obsidian with wiki-links, generates HTML reports. Integrates Your Business context (clients, projects, CRM). Triggers on /process command or daily 21:00 cron.
autograph
Schema-as-code enforcement for any Obsidian vault. Zero hardcoded domains. Use when creating vault cards, checking vault health, running schema compliance, deduplicating entities, generating MOC indexes, running decay cycles, bootstrapping a vault, fixing wikilinks, finding orphans or backlinks, extracting entities…
stop-experiment
Stop the autoresearch optimization loop cleanly. Disarms the launch-armed process watcher, cancels pending wakeups, reaps orphan workloads on the project's cluster pool, files any missing experiment pages, runs lint, and appends a clean-shutdown marker to wiki/log.md. Invoke at the end of an autoresearch session.
ingest-source
Ingest a paper, article, blog post, talk, or documentation page into the wiki. Encodes SCHEMA's INGEST-SOURCE operation as a callable skill. Reads the raw source (PDF/HTML under raw/sources/ or fetched from URL), confirms emphasis with the human before writing, files wiki/sources/ - .md, updates connections…