181 skills · sorted by relevance
Manage mathematical proof projects across Codex runs, keeping sources, attempts, and unresolved steps organized. Prepare manual GPT Pro handoffs for difficult steps, then check returned arguments and edit the proof for clarity.
Use Meta's Prompt Guard to detect malicious instructions in user prompts and external data. This skill guides input filtering, batch checks for retrieved documents, detection threshold adjustments, and splitting long text for inspection.
Classify failures in CI, automated checks run during development, to distinguish code regressions, temporary infrastructure problems, security test failures, and inconsistent results. Record each failure's category, suggested next action, and attempt count in a structured report.
Convert PDF, PowerPoint, Excel, and Word documents into structured Markdown for search and information extraction. Outputs include document metadata, a parsing confidence score, and the parser used, with fallback parsers available when earlier attempts fail.
Diagnose connection problems with government websites by checking SSL certificates, which verify website identity, and proxy settings. Follow structured troubleshooting steps, record failed attempts, and explore alternative APIs or data sources when access remains blocked.
Extract text from difficult PDFs by trying several libraries and command-line tools in sequence, then using OCR, which reads text from images. Record failed attempts and check that extracted content meets the task requirements.
Extract text from PDFs by checking each tool’s output and switching methods when needed. Guides agents from file reading to pdftotext and PyMuPDF, then checks readability, missing sections, and records extraction failures.
Download PDFs, check that the files contain valid content, and try several text extraction tools in sequence. The workflow helps detect downloaded error pages, insufficient text, and extraction failures while recording diagnostic details.
Assess Docker, containerd, and Podman environments for weaknesses that could expose the host system. The skill organizes checks for excessive permissions, exposed runtime sockets, and known vulnerabilities, alongside indicators defenders can monitor.
Hunt for signs of compromise in web and application logs without an initial alert. Establish normal activity, record investigative hypotheses, check source code, and reproduce issues locally before producing incident records and a report.
Test whether an AI agent reveals another user's data or performs actions beyond the caller's permissions. Dialogue probes cover access across users or organizations, admin functions, and attempts to gain higher privileges without authorization.
Tests whether user messages can override an AI agent’s instructions or impersonate a higher authority. It checks for exposed internal instructions and changed behavior, then reports confirmed findings with prompts and responses as evidence.
Guide role-specific work in LoopX auto-research, from defining research conditions and proposing hypotheses to experiments, evaluation, and adoption or retirement decisions. Keep evidence, failed attempts, task handoffs, and stopping conditions tied to each worker’s assigned role.
Inspect Android apps for embedded credentials, internal API addresses, and components exposed to other apps. The skill guides APK acquisition, code inspection, and runtime traffic analysis to support a mobile security assessment.
Review multi-factor authentication flows for skipped checks, reused one-time codes, weak backup paths, and verification races. It focuses on proving access to protected resources rather than relying on interface behavior or response codes alone.
Turn a recently solved research problem into searchable Markdown documentation. Capture symptoms, failed attempts, root causes, working code, and prevention checks, then organize the solution by category and link it to related records.
Design, debug, and tune OpenRouter API integrations. It checks model capabilities, reasoning settings, output limits, and fallback routes, then validates results against quality, cost, privacy, and performance requirements, including failed attempts.
Review progress while solving a math problem by organizing previous attempts, checking intermediate results, and assessing growing complexity. The skill recommends whether to continue, adjust the approach, switch strategies, or tackle a smaller problem first.
Search previous work records to recover useful decisions, successful approaches, and failed attempts. Find context in handoffs, plans, and build or test histories, then use those findings when tackling similar tasks or recurring errors.
Choose LLM settings for theorem proving, code generation, and creative exploration. It provides examples for output limits and response variety, plus proof planning prompts and a method for generating multiple proof attempts and selecting the best candidate.
Route requests to commit, push, or save Git changes through a dedicated /commit skill. The workflow removes Claude attribution, records attempted approaches in reasoning.md, clears build attempts, and pushes changes when requested.
Guide authorized testing of login, authorization, and sessions, including passwords, JWT tokens, OAuth, and multifactor authentication. The skill helps define boundaries, select alternative investigation paths, and document evidence needed to support findings or report unsuccessful tests.
Guide authorized web cache poisoning assessments, where a cache stores an attacker-influenced response. The skill defines test boundaries, suggests alternative investigation paths, and specifies evidence needed to report a finding or document unsuccessful attempts.
Guide authorized SQL injection testing, where input may change database queries. Define scope, choose alternative investigation paths, and organize requests, responses, and impact evidence, including reports that document unsuccessful checks without claiming a vulnerability is absent.