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 chgagne/claude-skills-research --skill self-skillgit clone --depth 1 https://github.com/chgagne/claude-skills-researchWrote 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/chgagne/claude-skills-research/self-skill)<a href="https://agentmods.dev/skills/chgagne/claude-skills-research/self-skill"><img src="https://agentmods.dev/badge/skills/chgagne/claude-skills-research/self-skill/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/chgagne/claude-skills-research/self-skill"><img src="https://agentmods.dev/badge/skills/chgagne/claude-skills-research/self-skill.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 | $0.00136 | $0.04381 |
| Opus 5.5 | $0.00054 | $0.01752 |
| Sonnet 5.5 | $0.00027 | $0.00876 |
| Haiku 4.5 | $0.00014 | $0.00438 |
Grade A, and why
self-surveying-literature 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 6d 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 — 268 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Self-surveying literature: what should this draft cite but doesn't?
Act as the best-read reviewer the paper will get. The deliverable is a short, verified, prioritized list of missing papers. Each one comes with the reason a reviewer would raise it and where it belongs in the draft. It is not a dump of search hits. Missing one paper that threatens the novelty claim costs the authors more than ten marginal "could also cite" suggestions, so put recall effort into finding close prior work and put precision effort into what you report.
Non-negotiables
- No paper from memory. Every reported paper must come from an API record retrieved in this
session (S2, arXiv, DBLP, Crossref). Memory is welcome for deciding what to look for. If you
think "Smith et al. did this around 2019", run
s2.py match "<title>"or search for it, and use the retrieved metadata. - Read before you claim. Read the abstract of every paper you report. Before calling a paper a novelty threat, read its method section in the full text. A title alone is never enough.
- Make sure it is really missing. The draft may already cite another version (arXiv vs. proceedings, journal extension, different title, odd bib key). The scripts dedupe by IDs and fuzzy titles. Before reporting, also grep the .bib for the first author's surname and the year.
- Respect the time frame. Prior work means work available before the draft's cutoff. Report later or concurrent work separately and label it as such.
- Don't edit the user's .tex/.bib unless asked. Write everything to the work directory.
Setup
S=/absolute/path/to/self-surveying-literature/scripts # this skill's scripts dir
W=./related-work-audit # work dir (created on demand)
Scripts use only the Python 3 standard library. The S2 key is read from
~/.config/scholarly/s2_key. S2 calls are throttled to about 1/s and arXiv calls to 1 per 3 s, so
large expansions take minutes. Run long commands in the background or in sequence, not in parallel
(parallel runs would break the rate limit). Every discovery command with --work $W appends to
$W/pool.jsonl and logs queries to $W/queries.tsv. Each script has --help.
What ships with it
5 files 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.
- 6d ago First seen · 268 lines · 136 tokens per session scan A b0539b5b2fea
self-surveying-literature is a skill published in the GitHub repository chgagne/claude-skills-research (5 stars, last pushed 2d ago), licensed MIT. It adds 136 tokens to every session and 4,381 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-10-10.
Other skills, from other repositories
paper-writer
Medical/scientific paper writing workflow skill. Manages the full pipeline from literature search to submission-ready manuscript. Creates and manages a project directory with IMRAD-format section files, literature matrix, reference management, and quality checklists. Supports both English and Japanese papers.…
food-research
Run a comprehensive, multi-source literature and evidence-synthesis workflow for food & nutrition science. Use when the user wants to research a food/nutrition topic in depth, do a literature review, build an evidence brief, screen and synthesize many sources, verify citations, or scope a systematic review.…
food-paper
Multi-subagent manuscript system for food & nutrition science covering the whole research process: understand the field, frame research questions, curate and analyze data, run statistics, build figures and tables, construct the discussion, draft, polish, and self-review — journal-aware throughout. Includes a…
food-pipeline
Master orchestrator for the whole food & nutrition research-to-publication workflow. Coordinates the specialist skills — each with its own subagent set — into one governed path: journal selection, research (food-research / food-deep-research), writing & analysis (food-paper), figures (food-figure), peer review…
food-review
Multi-reviewer peer-review system for food & nutrition manuscripts. Simulates an editorial panel — a coordinating editor, three domain reviewers (methodology/statistics, domain/novelty, integrity/ethics), and a devil's advocate — plus a formatting-compliance check against the target journal (APA 7.0 by default, or a…
food-deep-research
General-purpose deep research that produces a fully written, source-validated literature review on any question: scope it, design the method, discover and screen sources by journal ranking, validate every source, extract and verify evidence, synthesize, stress-test, then write and format the review (APA 7.0 by…