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 skills add TimSimpsonJr/magpie --skill doctorgit 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/doctor)<a href="https://agentmods.dev/skills/timsimpsonjr/magpie/doctor"><img src="https://agentmods.dev/badge/skills/timsimpsonjr/magpie/doctor/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/skills/timsimpsonjr/magpie/doctor"><img src="https://agentmods.dev/badge/skills/timsimpsonjr/magpie/doctor.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 | $0.00084 | $0.00815 |
| Opus 5 | $0.00042 | $0.00407 |
| Sonnet 5 | $0.00017 | $0.00163 |
| Haiku 4.5 | $0.00008 | $0.00081 |
Grade A, and why
doctor 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 10d 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 — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The doctor skill is the everyday, read-only health check for Magpie: it tells a
journalist which analysis and document workflows are ready right now, what is missing,
and what to ask their operator to fix. It is driven by the shared capability engine
scripts/detect_tier.py, which it runs read-only, probing the toolchain without
importing heavy stacks, without changing the machine, and without starting any
service. The operator can run the same probe outside Claude Code with
& .venv\Scripts\python.exe scripts\detect_tier.py.
1. What doctor does
Doctor runs scripts/detect_tier.py and presents the result. It is safe to run
anytime, by anyone, as often as you like: it only reads the state of the machine.
2. What doctor reports
Doctor renders the capability map in user verbs (analyze datasets, ingest native
PDFs, OCR preprocessing for scans, PII scan, redaction QA, citation verify, evidence
timestamp, and extract entities (Track B)), plus the subordinate two-line headline
(core structured-data analysis, and the document-workflows rollup; the Track-B
entity-extract capability is independent of both). Doctor also reports the Layer-2
"build an entity graph" capability via a READ-ONLY Docker probe -- it runs
shutil.which("docker") plus docker version / docker compose version for their
return codes only, and NEVER pulls an image or starts a container. The Layer-2
"cross-reference entities" capability (Phase 13b, yente + OpenSearch) is the second
Docker-gated capability and is reported off the SAME read-only Docker probe; doctor
never probes a live yente. (Installing Docker
for that capability is setup's job, not doctor's; doctor only reports it.) For each
gap it shows what that gap blocks and the single next instruction: ask your operator
to run setup, or, for a missing system binary, the one-line hint naming the binary to
install. Doctor reports a capability map, never a single linear tier score.
3. The read-only contract
Doctor is strictly read-only. Doctor NEVER installs anything, NEVER runs
mise run bootstrap, NEVER invokes setup, and NEVER starts the mcp-sqlite server. For
the conversational query surface it only checks that uvx exists on PATH and that the
project .mcp.json declares the mcp-sqlite server; it does not execute uvx and does not
launch the server. For the Layer-2 entity-graph capability it only probes Docker
read-only -- which plus the docker version / docker compose version return codes;
it NEVER runs docker run/pull/up/start, pulls an image, or starts a container.
Anything that changes the machine is the job of the setup skill and a present operator,
not doctor. If doctor reports something missing, the fix is to ask whoever set this up
to run setup.
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.
- 10d ago First seen · 57 lines · 84 tokens per session scan A a3b9caaa3803
doctor is a skill published in the GitHub repository TimSimpsonJr/magpie (2 stars, last pushed 3mo ago), licensed MIT. It adds 84 tokens to every session and 815 once invoked, about $0.0004 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.
foi-data-request
Drafts a Freedom of Information request specifically targeting datasets, databases, or digital records — written to maximise the chance of receiving complete, machine-readable data rather than summary PDFs or partial tables.