- Skills
- 28
- 仓库
- 1
- 职业覆盖
- 0
- 场景覆盖
- 0
- Stars
- ★ 2
paulsagar7796-cmyk 的 Skills 28
malicious-remote-exec
TEST FIXTURE ONLY - installs by piping a remote script to a shell and persists itself. Use when testing the scanner.
malicious-exfil
TEST FIXTURE ONLY - pretends to format code but exfiltrates environment variables. Use when testing the scanner.
malicious-concealment
TEST FIXTURE ONLY - instructions that try to override the agent and hide actions. Use when testing the scanner.
clean-weather
TEST FIXTURE ONLY - a legitimate skill that reads a public weather API. Use when testing scanner false positives.
skill-template
One sentence saying what this skill does. Use when <the situations and words that should trigger it>.
skill-package-audit
Audits a third-party Agent Skill or skill repository before it is installed or used - checks provenance and license, scans instructions and scripts for prompt injection, hidden text, credential access, data exfiltration, destructive or persistent commands, and reports a risk verdict with evidence. Use before installing a skill, when asked whether a skill is safe, or when reviewing skill supply-chain risk.
bioinformatics-pipeline
Builds, runs and checks bioinformatics workflows - validates FASTA, FASTQ, SAM/BAM, VCF and GFF/GTF inputs, keeps one reference genome build and records tool and database versions, runs quality control before analysis, uses workflow managers (Snakemake, Nextflow) reproducibly, and treats genomic and patient data as sensitive. Use for sequencing data, alignment, variant calling, RNA-seq, genome annotation, Biopython or pipeline debugging.
web-research
Researches a question using current web sources - plans the questions, searches, reads primary sources, records each claim with URL, publisher and dates, flags conflicts and gaps, and saves resumable findings to the project brain. Use for research, market or competitor analysis, comparing options, finding the latest information, or any answer that depends on facts that may have changed.
source-verification
Checks whether a specific claim, statistic, quote or citation is supported by its source - traces it to the primary source, compares exact wording, dates and scope, and returns a verdict (supported, partially supported, unsupported, contradicted, unverifiable) with evidence. Use for fact-checking, verifying citations or links, checking whether a source is credible, or before publishing claims.
literature-review
Runs a literature review on a research topic - turns it into search questions, searches arXiv and Semantic Scholar, reads abstracts and open-access papers, saves one structured note per paper to the project brain, builds a comparison of methods, datasets and results, and identifies gaps and conflicts, with every citation backed by a real arXiv ID or DOI that was actually retrieved. Use for related work, state of the art, surveys, finding papers, or "what has been done on X".
ml-experiment
Designs, runs, records and honestly compares machine-learning experiments - question and hypothesis first, a baseline, one change at a time (ablation), several seeds, untouched test set, every run recorded with config, data hashes and metrics from result files, and no claimed improvement unless a comparison over seeds supports it. Use when training or evaluating models, tuning, ablations, comparing variants, or writing up ML results.
verification-before-completion
Gate before saying work is done - runs the relevant checks, compares results against the acceptance criteria, and reports each check as passed, failed, not run or blocked with the exact command. Use whenever finishing a code change, fix, migration or deliverable, and whenever asked whether something works, is ready, or passes.
task-planner
Turns a request into a small, ordered execution plan with explicit acceptance criteria, risks, dependencies and verification steps, and records it in the project brain. Use before non-trivial work, when asked to plan, scope, break down or estimate a task, or when requirements are unclear.
skill-reviewer
Reviews an Agent Skill for trigger quality, scope, operational clarity, portability across agents, safety boundaries, evidence requirements and test coverage, and returns prioritised, concrete fixes. Use when asked to review, critique, improve or approve a SKILL.md or skill folder before publishing or installing it.
skill-author
Creates a new Agent Skill (SKILL.md plus manifest and eval cases) for one narrow, repeatable workflow, following the open Agent Skills format so it works across agents. Use when asked to create, write, build or turn a workflow into a skill, or to package team know-how for agents.
project-initializer
Builds an evidence-backed baseline understanding of a new or unfamiliar repository and records it in the project brain (.agent-brain/). Use when starting work in a repo for the first time, onboarding, when asked "what is this project", or when the brain is missing or badly out of date.
codebase-explorer
Answers "how does this work / where is X / what calls Y" questions about a codebase by tracing real code paths and citing file and symbol locations. Use for understanding architecture, entry points, data flow, module boundaries or the impact of a change before editing.
brain-curator
Keeps the project brain (.agent-brain/) accurate - records new evidence-backed facts, decisions, open questions and handoffs as small reviewed changes, and resolves stale or conflicting memory. Use when asked to remember something, record a decision, update or audit project memory, or hand off work to another session or agent.