Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/Mexregkan/claude-for-researchersnpx agentmods add skills/mexregkan/claude-for-researchers/check-pipelineWrote 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/mexregkan/claude-for-researchers/check-pipeline)<a href="https://agentmods.dev/skills/mexregkan/claude-for-researchers/check-pipeline"><img src="https://agentmods.dev/badge/skills/mexregkan/claude-for-researchers/check-pipeline/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/mexregkan/claude-for-researchers/check-pipeline"><img src="https://agentmods.dev/badge/skills/mexregkan/claude-for-researchers/check-pipeline.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00106 | $0.01035 |
| Opus 5 | $0.00053 | $0.00517 |
| Sonnet 5 | $0.00021 | $0.00207 |
| Haiku 4.5 | $0.00011 | $0.00103 |
Grade A, and why
check-pipeline scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 13d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/check-pipeline — verify a code matches its pipeline doc
Pipeline docs drift when the code changes: a function gets renamed, a cell is inserted (shifting all
later cell numbers), a data file is dropped, an I/O contract changes. This skill compares a
Pipeline/*.md against the current source and reports every mismatch, so the map stays trustworthy.
When to use
- Right after editing a documented code.
- Before relying on a pipeline doc you didn't just write.
- On request: "is the pipeline still accurate?", "check the codes match the pipelines".
- Across the board (no argument): check every
Pipeline/*.mdagainst its code.
Method
1. Pair each pipeline doc with its code. The doc's header names it ("Code: path/to/"). Read the header, or scan the folder if checking everything.
2. Dump the code to readable text (same helper the writer uses):
python3 .claude/skills/write-pipeline/dump_code.py "$SCRATCH" "path/to/<file>"
3. Extract the doc's checkable claims and verify each against the dump:
- Symbols (every name in the "Key symbols" table and in prose code-spans): does it still appear
in the code?
grep -F 'symbolName'the.full.txt(or source). Flag any that are absent (renamed/removed) — the highest-value drift signal. - Cell numbers: for each
| Symbol | Cell N |row, confirm the symbol's definition is at/near cell N in<base>.outline.txt(±2 tolerates minor insertions; larger shifts = flag "cell numbers stale, N inserted/removed above"). If many rows are off by a constant, an insert/delete happened — report the offset and the pivot cell. - Data files: every input/output file the doc says is loaded/written — does the code still
read/write it, and does the file exist on disk (
ls)? - Upstream/downstream links: the "Inputs/outputs" section names producers/consumers — spot-check
those still hold (e.g. the doc says stage B consumes symbol
Xfrom stage A; grep thatXis still built the way the doc claims). - Copied-layer coupling: if the doc says "this section is copied verbatim from
<other code>", diff the copied definitions against the source — silent divergence there is a real correctness risk, not just doc drift.
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.
- 13d ago First seen · 69 lines · 106 tokens per session scan A 8492551cb278
check-pipeline is a skill published in the GitHub repository Mexregkan/claude-for-researchers (52 stars, last pushed 9d ago), licensed MIT. It adds 106 tokens to every session and 1,035 once invoked, about $0.0005 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 skills, from other repositories
review-response
Systematic reviewer response workflow: parse comments, classify by severity, develop response strategy, write structured rebuttal. Use when asked to 'write rebuttal', 'respond to reviewers', 'draft review response', or 'handle R&R'.
test-iterate-loop
Autonomously diagnose a codebase, apply minimal fixes, and rerun tests until they pass or a real blocker is reached. Use when the user explicitly requests an iterative fix-until-green loop across Python, R, Julia, or HPC workflows.
postmortem
Deliver a structured post-mortem after incidents, mistakes, or stuck sessions. Use when the user requests a structured post-mortem after incidents, mistakes, or stuck sessions.
code-archaeology
Recover the structure, intent, and lineage of old code, data, or analysis files. Use when inherited or dormant research code must be understood before it is changed. Not for a quality review of already-understood code.
devharness
Drive and debug a running app via the devharness MCP server - launch or attach to Chrome and Node.js, set breakpoints and logpoints, inspect call stacks and variables, watch console and network, manage dev servers, replay any earlier tool call by its history index, and record reproduction sequences that verify a fix.…
latex-rescue
Diagnose and fix LaTeX compilation errors. Handles undefined control sequences, missing brackets, math mode violations, package conflicts, undefined references, and environment mismatches.