sofer
43 verified skills172 total stars
agent-config-audit
Check whether Claude and Codex have equivalent access to shared agent resources, skills, hooks, plugins, MCP servers, permissions, startup behaviour, and provider-specific adapter config. Use when comparing agent environments, debugging missing capabilities after restart, or deciding whether to symlink a resource or configure a runtime.
tools4
init-override
Create a symlink from AGENTS.md to CLAUDE.md in the current project directory. Use when the user wants to initialise a project to use AGENTS.md as the memory file instead of CLAUDE.md.
data-ai4
address-review
Read review comments on a GitHub PR, address them by fixing code, and reply to each comment. Use when a PR has received review feedback that needs to be acted on.
development4
anonymise
Anonymise CSV files by removing personal identifying information and adding datetime stamps. Use when user wants to process a new CSV file or strip PII from data.
development4
calendar-fetch
Fetch upcoming calendar events from Google Calendar
development4
design
Plan architecture and components for a single user story. Use after spec phase to create a technical design before implementation. Produces design documents that guide stubs and implementation.
documentation4
contact-research
Research a person given their name and email, returning a brief profile with role, company, LinkedIn summary, and previous interactions
data-ai4
design-review
Validate design before implementation. Use after design phase to ensure architecture meets standards, spec requirements, and quality criteria. Gates progression to stubs phase.
testing4
distribution-channel
Identify and evaluate distribution channels for the ML engineer apprenticeship
data-ai4
distribution-strategist
Discuss and evolve distribution strategy for the ML engineer apprenticeship
data-ai4
email-voice-draft
Draft an email reply in the user's voice using their business context and Notion knowledge
data-ai4
gmail-fetch
Fetch recent emails from Gmail with sender, subject, body, and thread context
data-ai4
intent
Interactive conversation to clarify user intent step-by-step before creating specifications or solutions. Use when requirements are unclear or user has a new idea to explore.
testing4
meeting-prep
Check calendar, research attendees, and produce a briefing document for each upcoming meeting
testing4
monitor
Observe application health and gather feedback after deployment. Use to validate success criteria, collect metrics, and feed issues back to the backlog. Closes the feedback loop.
testing4
postgres-query
Query a Postgres database (read-only) and return structured results
data-ai4
readme-review
Review the project README to understand what it does and suggest a recommended next step. Use when starting work on an unfamiliar project.
documentation4
refactor
Clean up implementation code while keeping tests green. Improve code quality without changing behaviour. Use after implement phase to polish code before review.
development4
test
Write tests against stubs following TDD red phase. Tests should fail initially (0% pass expected). Use after stubs to define expected behaviour before implementation.
testing4
commit
Git workflow operations with Conventional Commits. Supports subcommands - branch (create feature branch), commit (stage and commit changes), pr (create pull request), merge (merge PR). Automatically available when the current directory is a git repository. Use when user needs git operations during development workflow.
development4
implement
Write code to pass all tests (TDD green phase). Expect 100% test pass rate. Use after tests are written to implement the actual functionality.
development4
feature
Implement a feature using TDD. Works standalone (user describes what to build) or as part of the orchestrated pipeline (story and context provided). Handles complexity scaling, design, stubs, tests, implementation, migrations, refactoring, and review. Use when building a feature in an existing codebase.
development4
orchestrate
Coordinate full SDLC pipeline for software projects. Manages backlog, spawns phase agents with minimal context, gates transitions, and handles checkpoints. Use when starting a new project, picking up the next story, or resuming pipeline work.
testing4
reconcile
Analyse and resolve divergences between SDLC manifest and actual development state (git commits, branches, uncommitted files). Use when manifest and reality have drifted apart, or before picking next story to ensure clean state.
development4
requirements
Transform intent into user stories with acceptance criteria. Use after intent clarification to build a prioritised backlog. Produces stories in standard format ready for spec phase.
development4
spec
Define contracts, schemas, and behaviours for a single user story. Use after a story is selected from the backlog to create a detailed specification before design. Produces testable specifications.
testing4
stubs
Create interfaces, type definitions, and stub implementations that compile but are not functional. Use after design-review approval to establish code structure before writing tests. Enables TDD red-green-refactor cycle.
development4
problem-statement
Turn a vague or underspecified request into a self-contained problem statement. Use when the user has a rough idea, when a request would fail if handed directly to an agent, or before non-trivial work that needs shared understanding.
testing4
skill-usage
Record substantive skill use in an append-only local log. Use after choosing or invoking a non-system skill for real work, when a skill is inspected but not used, or when a skill fails to apply. Do not use for routine system skills or incidental file reads.
testing4
learning
Append a one-line learning to ~/.agents/learning-log.md. Use when the user types /learning, or when something genuinely worth remembering surfaced during work and the user confirms it should be captured.
data-ai4
decompose
Break a clear goal, plan, or project into independently executable pieces with dependencies and verification criteria. Use when work is too large for one coherent implementation slice or when delegation needs clean ownership boundaries.
development4
code-review
Review code changes for bugs, security issues, regressions, test gaps, and fit with the stated goal. Use for PRs, branches, commit ranges, staged changes, or files, especially before committing or shipping platform work.
development4
constrain
Define guardrails for significant work before delegating or implementing it. Use when failure modes, ownership boundaries, platform safety, or technically-correct-but-wrong outcomes matter.
data-ai4
evaluate
Define evaluation criteria, tests, or quality checks for a deliverable, implementation, workflow, or recurring agent task. Use when defining done, improving a test harness, or checking whether output meets its specification.
testing4
deploy
Ship code to target environment. Use after code is merged to deploy to staging or production. Handles build, artifact creation, and deployment orchestration.
development4
email-drafter
Fetch unread emails, triage them, draft replies in the user's voice, and save as Gmail drafts for human review
data-ai4
meeting-context
Determine why a meeting is happening and what should be discussed
tools4
curriculum-module
Create a learning module for the platform. Teaches a topic collaboratively, structures it into chapters with steps and questions, then generates upload-ready markdown. Use when the user wants to create curriculum content.
content-media4
draft-post
Guide the user through creating a blog post draft for Substack (200-500 words) and LinkedIn (100-300 words). Use when user wants to draft a post or share work.
documentation4
gmail-draft
Create a draft email in Gmail from provided email data
data-ai4
notion-query
Search and retrieve content from a Notion workspace
content-media4
user-test
Reformat code-review manual test script for human or agent execution, run user testing checkpoint, and record pass/fail results. Gates progression to commit:pr.
development4
end-to-end-verification
Verify public-facing user flows before release or production push, especially flows involving email, authentication links, multi-step forms, generated links, or records that users will interact with.
testing4