SQLite best practices for Python -- connection setup, WAL mode, foreign keys PRAGMA (must be set per connection!), context managers for transactions, parameterized queries, row_factory, executemany, connection management, indexes on FKs, and money as INTEGER cents. Use when building or reviewing any Python app with SQLite, when you see sqlite3 imports, or when setting up a new local database in a FastAPI/Flask project.

Tests memory writes, confirms read-back accuracy, and validates retrieval success to ensure saved information can actually be recovered. Use when you need to verify memory was saved correctly, check if stored data can be retrieved, confirm a memory entry is discoverable, or escalate when saved information appears lost or corrupted. Covers write confirmation, read-back comparison, retrieval smoke testing, and failure escalation.

Create terminal-based presentation slides using presenterm's markdown format. presenterm renders markdown files as slides in the terminal with themes, code highlighting, images, column layouts, speaker notes, and more. Use this skill whenever the user wants to create a presenterm presentation, terminal slides, markdown slides for presenterm, or mentions "presenterm" in any context. Also trigger when the user asks for "terminal presentation", "markdown presentation", "slide deck in markdown", or wants to convert content into presenterm format. Even if the user just says "create a presentation" and they have used presenterm before or the context suggests terminal-based slides, use this skill.

Resolve pull request feedback by critically assessing each item, fixing source files (not generated files), regenerating derived artifacts, running lint/format, committing, pushing, and replying on the appropriate GitHub thread. Use when the user asks to address PR feedback, respond to reviewer suggestions, fix issues from code review, or resolve GitHub review comments, PR issue comments, or review summaries.

Runs a structured 10-check pre-submission checklist against an open-source contribution before the contributor opens a pull request. Use when the user has written code for an open-source project and needs to prepare a PR, submit a contribution, or verify readiness. Triggers on "submit a PR", "open a pull request", "prepare the contribution", "ready to merge", "check my pull request". IMPORTANT — run this AFTER code is written but BEFORE submission. Checks: AI policy compliance and disclosure (including voluntary disclosure when no policy exists), diff size and focus, PR template, code style, commit conventions, tests, changelog and metadata, legal requirements (DCO/CLA), agent artifacts, slop detector patterns, and human ownership verification.

Analyzes project contribution guidelines, identifies the right venue (pull request, issue, discussion, RFC/KEP/DEP), checks issue metadata (claims, assignments, labels), searches for prior rejected attempts, and drafts proposals formatted to project templates. Use when the user wants to contribute to an open-source project, fix a bug, submit a PR, improve or refactor code, asks where to submit a change, or needs help choosing between PR/issue/discussion/RFC. Triggers on "fix this issue", "submit a PR", "refactor this", "improve this code", "open a pull request". IMPORTANT — run this AFTER recon and BEFORE writing code to verify the right venue and check for prior attempts.

Analyze an open source project's contribution norms, AI policy, conventions, and recent PR history before writing any code. Use when the user wants to contribute to an open source or GitHub project, fix a bug, submit a pull request, open a PR, make a contribution, or asks about contribution guidelines. Triggers on phrases like "fix this bug", "submit a PR", "contribute a fix", "open a pull request", "help me contribute", "how do I contribute", "what are the rules for this OSS project". IMPORTANT — run this BEFORE writing any code for an open source project.

Turn a Kibana JSON log export into a runnable pytest suite using the secure-log2test CLI. Use when the user has a Kibana or Elasticsearch JSON export of API traffic and wants a regression suite from production logs, when extracting test cases from staging traffic, when scrubbing auth headers or secret-looking body fields before logs leave the laptop, when bridging Kibana-captured requests into a pytest-based suite for CI, when the user mentions Kibana logs, Elasticsearch JSON export, log-to-test conversion, log replay tests, auth header redaction, PII in logs, or regression tests from production traffic.

Apache Flink SQL, Table API, and UDF development for both OSS Flink and Confluent Cloud. Use when: (1) Writing Flink SQL queries (windows, joins, aggregations, MATCH_RECOGNIZE), (2) Building Table API pipelines in Java or Python, (3) Creating UDFs (scalar, table functions) for Flink, (4) Deploying Flink jobs to Confluent Cloud, (5) Converting between DataStream and Table API, (6) Troubleshooting Flink SQL errors. Covers windowing, event-time processing, watermarks, state management, and Confluent-specific patterns.