synapseradio
48 verified skills74 total stars
thinkies-detect-diminishing-returns
Detect when further effort yields little gain
data-ai2
ponder
Exploration skill for problems that need thinking before solving. Assesses a problem's shape and applies 1-5 exploration techniques in sequence — producing questions, reframings, structural maps, and new angles. Use when asked to "think about this", "help me explore", "what am I missing", "I'm stuck", "help me understand this problem", "ponder this", "what should I consider", "help me think through this", or when a problem needs reframing, assumption-surfacing, or systematic questioning before action. Also use when the user describes a vague idea, faces a decision fork, or expresses uncertainty about what they don't know.
testing2
thinkies-ask-respond
Structured Q&A that decomposes questions before answering
tools2
thinkies-argue-the-opposite
Stress-test a position by building the strongest counter-case
development2
prompt
Craft or refactor LLM instructions. Use when using the Agent or Task tool, when the user says "write a prompt", "prompt", "prompt an agent", or any similar request.
tools2
thinkies-audit-chain-of-thought
Tag reasoning steps by inference type
testing2
thinkies-check-soundness
Test synthesis for contradictions
testing2
thinkies-branch-possibilities
Generate fundamentally divergent directions from one starting point
tools2
ts-typeclasses
Implement typeclasses and their instances in TypeScript — Functor, Applicative, Monad, Traversable, Foldable — and the higher-kinded type (HKT) encoding they require. Use this skill whenever the user wants to give a custom data type typeclass instances; abstract over a type constructor or an unknown container `F`; encode higher-kinded types; work with `TypeLambda`, `Kind`, `HKT`, or `@effect/typeclass`; decide whether a typeclass is the right abstraction at all — including in a regular, non-FP codebase where it cuts against the surrounding style; or debug HKT-encoded code (inference failures, variance slots, unreadable `Kind` error messages). The skill uses Effect's HKT encoding as the substrate, but the subject is typeclasses in TypeScript generally, in any repo. It first checks whether a typeclass is warranted, then walks the brand/`TypeLambda` encoding with copy-ready templates and a catalog of standard typeclasses.
development2
thinkies-calibrate-confidence
Match certainty to evidence strength
data-ai2
thinkies-assess-current-knowledge
Map what's known vs assumed vs unknown
tools2
thinkies-detect-fallacies
Spot logical errors in reasoning
testing2
thinkies-excavate-assumptions
Surface unstated assumptions at multiple levels and rank them
tools2
thinkies-ideate
Generate and filter ideas into vetted options
tools2
thinkies-integrity
Verify epistemic integrity by aligning claims with evidence
testing2
thinkies-integrate-other-perspectives
Combine viewpoints into coherent whole
tools2
thinkies-invert-the-problem
Turn a problem inside out to reveal hidden structure
tools2
thinkies-consider-alternatives
Generate competing explanations for the same observations
tools2
thinkies-connect-domains
Import solutions from structurally similar problems in distant domains
data-ai2
thinkies-ask-what-breaks
Find defeaters that would break a conclusion
tools2
thinkies-evaluate-evidence
Assess how well evidence supports claims
data-ai2
thinkies-derive-first-principles
Strip convention to irreducible truths and rebuild
development2
thinkies-flip-assumptions
Test claims by forming contrapositive
testing2
thinkies-cite-sources
Track, validate, and cite external sources with working URLs
testing2
thinkies-decompose
Break a whole into parts at its natural joints
tools2
thinkies-find-leverage
Locate where a small change shifts the whole system
data-ai2
tree-of-thought
Systematic Tree of Thought reasoning for complex problem decomposition. This skill should be used when the user asks for "tree of thought", "systematic reasoning", "structured analysis", "multi-path evaluation", or wants to decompose a complex problem by generating and evaluating multiple solution paths before selecting the best approach. Applies four-phase reasoning: decomposition, generation, evaluation, and synthesis.
research1
apache-age
Apache AGE graph database extension for PostgreSQL — Cypher query syntax, agtype data types, hybrid SQL+Cypher patterns (CTEs, JOINs, parameter passing), graph schema modeling (vertices, edges, labels, properties), and performance optimization (indexing, EXPLAIN, batch loading). Use when writing Cypher queries, designing graph schemas, combining graph traversal with relational SQL, or optimizing AGE query performance — all within PostgreSQL via the AGE extension.
tools1
shell-testing
Write idiomatic BATS tests for bash and zsh shell scripts. Use when writing new tests, adding tests to existing shell projects, reviewing test quality, running tests after code changes, or when the user asks to "write tests", "add tests", "test this script/function", "run tests", or "create a test file". Also triggers on: .bats files, bats-assert, bats-file, shell test organization, zunit, or testing shell plugins/commands/functions. Covers BATS-core patterns, assertions, mocking, isolation, zsh-specific testing, and CI integration.
tools1
de-residency-advisor
Use when the user asks for help preparing for a German government appointment on visa, residency, or citizenship — Fiktionsbescheinigung, Aufenthaltserlaubnis, EU Blue Card, Approbation / Berufserlaubnis (non-EU doctors), Einbürgerung, the Einbürgerungstest, national visa interviews at a German embassy, residence permit renewals, family reunification, the Chancenkarte, or any Ausländerbehörde / LEA / Einbürgerungsbehörde / Standesamt appointment. Triggers: "Fiktionsbescheinigung", "Aufenthaltserlaubnis", "Blue Card", "Approbation", "Einbürgerung", "Einbürgerungstest", "LEA appointment", "Ausländerbehörde", "German residence permit", "German citizenship", "moving to Germany paperwork". Researches every answer live and cites every factual claim with a URL; never relies on training data for fees, thresholds, deadlines, or document lists. Speaks in a warm, conversational voice modelled on Erin Hannon: earnest, gently encouraging, never glib.
development1
team
Expert persona collaboration for structured problem analysis. This skill should be used when the user asks to "assemble a team", "team analysis", "expert panel", "persona collaboration", or wants multiple expert perspectives applied to a problem. Creates a team of 3-8 expert personas who collaboratively analyze problem spaces through phased conversation: situation analysis, team selection, discussion, reasoning, and optional implementation.
research1
communicate
Communicate ideas and information to others with purpose, clarity, and integrity. Ensure that artifacts are structured and phrased in a way that is mindful of their audience while avoiding AI slop patterns. Use when the user requests assistance with writing, commentary, or communication; when they point out AI slop, when they ask things like "help me say", "write for [a specific audience or context]", "polish [comments, sentences, artifacts]", or the like.
development1
bash-scaffold
Scaffold a production-grade bash script from a curated template. Use whenever the user asks to "scaffold a bash script", "start a new shell script", "write a shell script for X", "give me a bash script template", "bootstrap a .sh", "I need a maintainable bash script", or describes any task they intend to automate in bash from scratch — even when they don't explicitly say "scaffold".
testing1
communicate
Written communication skill with 16 techniques across 5 categories and 5 structured workflows. Use when asked to "improve my writing", "refine this prose", "make this clearer", "proofread this", "clean this up", "edit this", "check my writing", "make this as good as it can be", "full refinement", "polish this thoroughly", "help me write this", "draft this for me", "help me compose", "adapt this for a US audience", "translate the argument structure", "why does my writing feel foreign?", "help me say this", "I can't find the right words", "in [language] we say...", or when any writing improvement, composition, structural adaptation, or cross-lingual expression work is needed.
development1
cli-development
CLI development reference grounded in https://clig.dev. Use this skill whenever designing, reviewing, modifying, or specifying a command-line interface — drafting flags or subcommands, writing help text, choosing exit codes or error output, picking environment variables, choosing config file locations, naming a binary, deciding stdout vs stderr, planning interactivity or progress output, reviewing CLI source code in pull requests, writing CLI specs or RFCs, or doing any development, design, or QA-adjacent work that touches CLI surface. The skill provides a sitemap of clig.dev anchors plus local reference copies of every guideline section, and instructs the agent to pull live content from clig.dev before citing rules so recommendations stay current.
tools1
create-skill
Create research-backed, spec-compliant Agent Skills with cited references, workflow templates, and anti-hallucination guardrails. Use when asked to "create a skill", "make a new skill", "build a skill for X", or when designing specialized Claude capabilities.
development1
runbook
This skill should be used when the user asks to "create a runbook", "set up an autonomous loop", "implement this feature on a loop", "build this out autonomously", "start a hardening pass", "run a refactoring loop", "start a research sweep", "focus on X and loop", "runbook", "set up a cron loop", or when the user wants to decompose a task, initiative, or project into a steerable autonomous workflow. Also triggers on: FOCUS.md, autonomous loop, or any request to set up recurring agent-driven work — whether building, implementing, hardening, refactoring, or exploring.
development1
sequencer
Parse concise pipeline instructions into ordered task sequences and execute them with accumulating context. This skill should be used when the user asks to "run a sequence", "chain skills together", "pipeline these steps", "seq", types "/seq", or provides arrow-separated (via the arrow symbol), bulleted, or prose-described multi-step instructions that mix skills, agents, and natural-language tasks.
testing1
rabbit-hole
Multi-agent investigation pipeline for deep research questions. Use when a question requires searching across multiple sources (codebase, web, docs, academic), synthesizing findings, and validating citations. Triggered by: "rabbit hole [topic]", "deep dive into [topic]", or any complex research question where a single search would be insufficient. Readonly — never modifies files. Always cites sources. Always validates citations.
development1
skill-review
Evaluate Agent Skill directories against the Agent Skills Specification and quality criteria. This skill should be used when the user asks to "review a skill", "audit skill compliance", "check if a skill passes", "validate skill against spec", or when verifying a skill before publishing. Produces a pass/fail report with fix instructions for each failure.
testing1
visualize
Visualize data, concepts, relations, or diagrams. Produces browser-runnable HTML charts and markdown outputs that drop into pull requests, READMEs, tickets, and chat. Use whenever the user asks to visualize anything, make a chart or diagram, plot a trend, compare numbers, show a flow or timeline, or pick the right chart type.
development1
waypoint
Distributed navigation system for multi-file pipelines and processes. Creates compact, grep-friendly waypoint markers that orient readers within complex cross-file workflows — each file carries its own ID, role, adjacent nodes, and a pointer to the full map. Use when asked to "point me to", "where does X begin/end", "map this pipeline", "add waypoints", "trace this process", or when working with processes spanning many files that need discoverability without centralized documentation that drifts.
devops1
shape-up
Conversational software requirements elicitation producing shaped specifications. This skill should be used when the user asks to "write a spec", "create requirements", "spec this out", "what should we build", "define the requirements", "help me think through this system", "spec out a feature", "shape this project", "requirements gathering", "write a pitch", "help me figure out what to build", or when a user describes a software need that requires structured requirements before implementation begins.
development1
stax
Manages stacked Git branches and PRs using stax (v0.27.0). This skill should be used when the user asks to "create a branch", "push changes", "rebase", "submit a PR", "sync with main", "manage stacked branches", or when any git branch, rebase, push, or PR operation could be handled by stax. Provides stax-first routing so agents default to stax over raw git for all supported operations, plus composable workflows for stacked development.
development1
flix
This skill should be used when the user asks to "write Flix code", "translate to Flix", "set up a Flix project", "fix Flix compiler errors", "add Java interop in Flix", or when encountering .flix files or flix.toml. Context engineering runbook providing design constraints from Flix's 41 principles, stdlib-first discipline via api.flix.dev, and anti-hallucination guidance for a language with sparse training data.
development1
de-residency-advisor
Use when the user asks for help preparing for a German government appointment on visa, residency, or citizenship — Fiktionsbescheinigung, Aufenthaltserlaubnis, EU Blue Card, Approbation / Berufserlaubnis (non-EU doctors), Einbürgerung, the Einbürgerungstest, national visa interviews at a German embassy, residence permit renewals, family reunification, the Chancenkarte, or any Ausländerbehörde / LEA / Einbürgerungsbehörde / Standesamt appointment. Triggers: "Fiktionsbescheinigung", "Aufenthaltserlaubnis", "Blue Card", "Approbation", "Einbürgerung", "Einbürgerungstest", "LEA appointment", "Ausländerbehörde", "German residence permit", "German citizenship", "moving to Germany paperwork". Researches every answer live and cites every factual claim with a URL; never relies on training data for fees, thresholds, deadlines, or document lists. Speaks in a warm, conversational voice modelled on Erin Hannon: earnest, gently encouraging, never glib.
development1
scamper
Apply the SCAMPER creative thinking technique for structured ideation. This skill should be used when the user asks to "brainstorm ideas", "use SCAMPER", "creative thinking", or wants to explore a problem through systematic creative exploration.
data-ai1
cite
Generate APA-format citations from paper links. This skill should be used when the user asks to "cite a paper", "generate APA citations", "format references", or provides paper links (arXiv, DOI, conference URLs) for citation formatting.
development1