Agent skill

Epupp Dev

by PEZ in PEZ/epupp

Epupp architecture, development patterns, and system contracts.

MITAuto-check passedDevelopment

Install Epupp Dev

skills CLI
$ npx skills add PEZ/epupp --skill epupp-dev -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install PEZ/epupp epupp-dev --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/PEZ/epupp.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.github/skills/epupp-dev .claude/skills/epupp-dev && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
epupp-dev
GitHub stars
116
Token cost
~2.4k tokens
SKILL.md length
518 words
Files
5 (incl. references)
Skills in repo
5
Repo updated
First seen
Licence
MIT

At a glance

Epupp architecture, development patterns, and system contracts.

  • Modifying Epupp source code
  • SKILL.md covers Architecture, Principles, Uniflow (compact) and REPLs and Babashka, plus 5 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
  • Implementing features

What it does

Epupp Dev is an agent skill from PEZ/epupp. Epupp architecture, development patterns, and system contracts. Use when: writing or modifying Epupp source code, implementing features, fixing bugs, debugging extension behavior, working with Uniflow actions or effects, understanding message flows, injection or connection lifecycle, testing strategy, build pipeline, reviewing Epupp PRs, planning implementations, running tests or builds, REPL-driven development, or any coding task in the Epupp workspace. Contains architecture overview, Uniflow contracts, source…

Its SKILL.md is about 2.4k tokens, which your agent loads only when the skill is triggered. The skill folder holds 5 other files, including reference files (for example `references/decision-rules.md`, `references/message-protocol.md` and `references/state-contracts.md`).

It sits in Development, covering Test strategy and CI/CD. It works with Chrome Extensions. The repository describes itself as: A web browser extension that lets you tamper with web pages, live and/or with userscripts. The licence is MIT.

When your agent uses it

  • Modifying Epupp source code
  • Implementing features
  • Debugging extension behavior
  • Working with Uniflow actions

Example prompts

  • “/epupp-dev”

Requirements

  • Python 3

What it can do on your machine

Read from SKILL.md and the folder at commit 566ce34. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Epupp Dev loads about 2.4k tokens when it runs, and up to ~7.9k if it reads all its reference files. Until then it costs about 185 tokens; SKILL.md has 518 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~185
When it runs · the whole SKILL.md, loaded when a task matches
~2.4k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~7.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from PEZ/epupp at commit 566ce34, republished under its MIT licence (© PEZ). 518 words, ~2,425 tokens.

Download SKILL.mdSave it as .claude/skills/epupp-dev/SKILL.md (or your agent's skills folder). This skill also uses 4 other files; get the full folder from GitHub.
name
epupp-dev
description
Epupp architecture, development patterns, and system contracts. Use when: writing or modifying Epupp source code, implementing features, fixing bugs, debugging extension behavior, working with Uniflow actions or effects, understanding message flows, injection or connection lifecycle, testing strategy, build pipeline, reviewing Epupp PRs, planning implementations, running tests or builds, REPL-driven development, or any coding task in the Epupp workspace. Contains architecture overview, Uniflow contracts, source file map, testing patterns, delegation, and development principles. References hold deep state contracts and temporal sequences. IMPORTANT: Also load when PLANNING Epupp changes, not only at implementation time.

Epupp Development

Shared knowledge base for developing the Epupp browser extension. Loaded by any agent or mode doing real Epupp work.

Architecture

λ architecture_layers. editor_or_AI → nREPL_client → babashka_relay(browser-nrepl, ports_12345/12346) → background_service_worker(chrome.runtime) → content_bridge(ISOLATED_world, chrome.runtime ↔ postMessage) → ws_bridge(MAIN_world, postMessage ↔ WebSocket) → page_context(Scittle_REPL ↔ DOM) | six_layers | each_boundary ≡ trust_boundary ∧ serialization_point | ¬direct_editor_to_page | always_relayed

λ build_pipeline. src/.cljs → squint_compiler → extension/.mjs(ESM) → esbuild → build/.js(IIFE) | config: squint.edn | bb_squint-compile ≡ compile_check | bb_watch ≡ continuous | build/.css ∧ build/.html copied_from extension/ | format: IIFE | ¬ESM_in_browser_extension | define: EXTENSION_CONFIG from config/*.edn | separate_bundles: background.js popup.js panel.js content-bridge.js ws-bridge.js trigger-scittle.js userscript-loader.js

λ page_runtime. scittle ≡ SCI_in_browser | true_clojure_keywords | userscripts_run_in_page_context | full_DOM_access | async_via ^:async ∧ await | ¬top_level_await | libraries_injected_via :epupp/inject | bundled_scittle_plugins

λ injection_model. ensure_scittle! → check_first → inject_only_if_missing | ensure_bridge! → check_first → inject_only_if_missing | duplicate_injection → ¬error | ¬duplicate_state | check_scittle_fn → {hasScittle, hasWsBridge} | both_must_be_true

λ content_bridge_security. ISOLATED_world: content_bridge.cljs | ¬page_access | chrome.runtime_access | validates_message_source ∧ type via message_registry | MAIN_world: ws_bridge.cljs | page_access | ¬chrome.runtime_access | trust_boundary: ISOLATED ≡ trusted | MAIN ≡ untrusted

Principles

λ epistemology. assumptions ≡ enemy | benchmark > estimate | measure > guess | verify_before_stating_fix_locations | mark("needs investigation") when_uncertain | failure_to_read_docs ≡ #1_cause_of_mistakes

λ data_oriented. what_would_rich_hickey_do | data > objects | fn > class | immutable_core > mutable_shell | transform(data) > mutate(state) | REPL_first: test_functions ∧ explore_data ∧ validate_assumptions_before_coding | simple > easy | separate(data, behavior) | clear > clever

λ clojure. definition_order_matters | ¬forward_declares | almost_always ≡ poor_structure | plan_test_strategy: unit(structural ∧ contracts ∧ invariants) ∧ e2e(integration ∧ real_flows) | unit ∧ e2e ≡ complementary_tools | ¬sequential_phases

λ api_stability. user_facing_API ≡ commitment: manifest_keys ∧ epupp.fs ∧ REPL_behaviors | preserve_existing | break_only_when(cost_compat > cost_users) | prefer_clean_solutions | ¬contorted_shims | ¬special_case_branching

Uniflow (compact)

λ uniflow_sap. event_loop ≡ SINGLE_ACCESS_POINT for !state | actions: pure_fn(state, uf-data, [action & args]) → {:uf/db :uf/fxs :uf/dxs} | effects: receive(params_from_action) | ¬read(@!state) | ¬transitive_atom_access | ¬swap! ∧ ¬reset! outside_event_loop | ref: dev/docs/architecture/uniflow.md | full_contracts: references/state-contracts.md

λ action_naming. actions: :domain/ax.verb-noun | effects: :domain/fx.verb-noun | state: :domain/key-name | messages: "kebab-case-string"

REPLs and Babashka

λ available_repls. bb → babashka_REPL | scripting ∧ automation squint → squint_REPL | test_pure_functions_in_Node.js scittle-dev-repl → scittle_Dev_REPL | test_scittle_code_in_browser_like_env | develop_solutions_incrementally_in_appropriate_REPL

λ babashka_utilities. prefer_babashka_builtins > python ∧ shell ∧ external_tools | http_server: bb_test:server ∨ babashka.http-server | ¬python_-m_http.server | file_ops: babashka.fs | ¬shell(cp mv rm find) | process: babashka.process | ¬raw_shell_scripts | http_client: babashka.http-client | ¬curl | ¬wget

Testing

λ testing_pattern. unit_tests: vitest | src/test/.cljs → test/.cljs | test_pure_actions: call_handle-action → assert_return_shape | test_data_transforms: call_fn → assert_output | ¬test_effects_in_unit_tests | effects_tested_in_e2e e2e_tests: playwright_in_docker | e2e/*.cljs | test_full_flows: injection ∧ connection ∧ eval ∧ userscripts | bb_test:e2e → docker_build → playwright_run

λ truth_hierarchy. browser_page > e2e_test > unit_test > source > docs > assumption | browser_page ≡ ground_truth | where_scittle_actually_runs | ¬trust(green_tests_alone) → verify_in(real_browser)

λ debug_approach. 1_context: gather(failing_env ⊗ working_env) | what_data_differs 2_trace: data_flow(input → transform → output) | find_divergence_point 3_fix: root_cause(data_flow) | ¬symptom_patch | six_layer_trace: page_scittle → ws_bridge → content_bridge → background → bb_relay → editor | find_break_point

Show full SKILL.md (193 more words)Show less

Delegation

λ subagents. commit → summary_of_task(bigger_picture) | expert_git_agent research → context ∧ what_to_know ∧ report_structure edit → Clojure_editor | files ∧ linenumbers ∧ code ∧ instructions epupp-elaborator → user_prompt ∧ file_context ∧ session_context → refined_prompt | delegation ≡ intelligent | protect_context_window ∧ ensure_quality

Documentation

λ docs_by_task. read_before_starting_work | use(read_file): | understanding_system → architecture_overview → detailed_docs | unit_tests → dev/docs/testing-unit.md | UI(popup/panel) → reagami skill ∧ dev/docs/ui.md | state/events → dev/docs/architecture/uniflow.md ∧ state-management.md | messaging → dev/docs/architecture/message-protocol.md | injection/REPL → dev/docs/architecture/injection-flows.md | library_deps → dev/docs/architecture/library-namespaces.md | userscripts → dev/docs/userscripts-architecture.md | build/release → dev/docs/dev.md | finding_source → dev/docs/architecture/components.md | return_to_index_when_scope_changes

λ docs_reference. consult_when_relevant: | README.md → user_facing_overview | dev/docs/architecture/security.md → trust_boundaries ∧ CSP | dev/docs/architecture/build-pipeline.md → build_config_injection

λ docs_sync. epupp_repo ≡ source_of_truth | synced_files ≡ generated_artifacts | README.md → my-epupp-hq/docs/epupp-README.md | docs/repl-fs-sync.md → my-epupp-hq/docs/repl-fs-sync.md | docs/connecting-to-epupp.md → my-epupp-hq/docs/connecting-to-epupp.md | ¬edit(synced_copies) | treat_as_generated | ∀doc_updates: edit_here(epupp/) | human_runs(bb_docs-sync)_when_ready

Source File Map

λ source_file_map. src/background.cljs → service_worker | message_routing ∧ init src/popup.cljs → popup_ui | connection_controls ∧ script_list src/panel.cljs → devtools_panel | code_editor ∧ eval src/content_bridge.cljs → ISOLATED_world | message_validation ∧ relay src/ws_bridge.cljs → MAIN_world | WebSocket_relay ∧ page_comms src/trigger_scittle.cljs → page_injection | Scittle_loader src/userscript_loader.cljs → content_script | auto_injection src/event_handler.cljs → uniflow_engine | dispatch ∧ execute src/storage.cljs → chrome.storage_mirror | persist ∧ load src/config.cljs → build_config | dev/prod/test src/manifest_parser.cljs → EDN_manifest_parsing src/script_utils.cljs → script_normalization ∧ ID_generation ∧ library_classification src/scittle_libs.cljs → library_collection ∧ injection src/bg_ws.cljs → background_WebSocket_management src/bg_inject.cljs → content_script_injection src/dep_resolver.cljs → dependency_resolution | pure_resolver ∧ cycle_detection ∧ ext-dep_classification src/ext_dep.cljs → ext_dep_URL_validation ∧ async_fetch_cache src/bg_fs_dispatch.cljs → FS_message_routing ∧ optional_:uf/dxs_dispatch src/popup_actions.cljs → popup_uniflow_actions src/panel_actions.cljs → panel_uniflow_actions src/reagami.cljs → minimal_UI_rendering_library src/background_actions.cljs → background_uniflow_actions | re-resolution src/background_actions/*.cljs → background_action_modules

Deep References

Load as needed per task:

ReferenceWhen to load
state-contracts.mdAdding/modifying actions, effects, state shapes
message-protocol.mdDebugging message relay, adding messages
temporal-sequences.mdDebugging connection/injection flows
decision-rules.mdArchitecture decisions, "why does X happen"

© PEZ, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 4 other files (references) in .github/skills/epupp-dev of PEZ/epupp.

  • SKILL.md
  • references/decision-rules.md
  • references/message-protocol.md
  • references/state-contracts.md
  • references/temporal-sequences.md

Open the folder on GitHubat commit 566ce34

Compare with similar skills

Epupp Dev next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Epupp Dev compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Epupp Dev this skillPEZ/epupp116—~2.4kAutomated safety check: PassMIT
MAUI UI Test Shard Rebalancerdotnet/maui23k—~995Automated safety check: PassMIT
Automating Devopstelagod/code-abyss243—~422Automated safety check: PassMIT
PR Babysitteropeninterpreter/openinterpreter69k3 repos~4.2kAutomated safety check: PassApache-2.0
Robotics Testingarpitg1304/robotics-agent-skills368—~4.7kAutomated safety check: PassApache-2.0
CI Fixwarpdotdev/oz-skills825—~790Automated safety check: PassMIT

Similar skills

  • Official

    Splits a slow .NET MAUI UI-test category into method-level CI shards sized from historical Azure DevOps timings so each job fits a target duration.

    23k GitHub stars~995 tokensUpdated today
    Testing & QAAuto-check passed
  • Automating Devops

    telagod/code-abyss

    DevOps knowledge reference covering Git workflows, testing strategies, DevSecOps, release pipeline orchestration (release.yml, multi-arch images, cosign integration), CI/CD pipelines, database…

    243 GitHub stars~422 tokensUpdated 2 mo ago
    DevOps & CloudAuto-check passed
  • PR Babysitter

    openinterpreter/openinterpreter

    Watches an open GitHub pull request until it merges, handling review comments, diagnosing CI failures and retrying flaky checks along the way.

    69k GitHub starsUsed in 3 repos~4.2k tokens
    DevelopmentAuto-check passed
  • Robotics Testing

    arpitg1304/robotics-agent-skills

    Testing strategies, patterns, and tools for robotics software.

    368 GitHub stars~4.7k tokensUpdated 1 mo ago
    Testing & QAAuto-check passed
  • CI Fix

    warpdotdev/oz-skills

    Diagnose and fix GitHub Actions CI failures. An agent skill from warpdotdev/oz-skills.

    825 GitHub stars~790 tokensUpdated 1 mo ago
    DevelopmentAuto-check passed
  • Fix CI

    ccusage/ccusage

    Diagnoses and fixes failing GitHub Actions checks with gh. An agent skill from ccusage/ccusage.

    19k GitHub stars~487 tokensUpdated today
    DevelopmentAuto-check passed

More from PEZ/epupp

  • Epupp Squint

    PEZ/epupp

    Epupp-specific Squint source conventions for src/. An agent skill from PEZ/epupp.

    116 GitHub stars~814 tokensUpdated 13 days ago
    Auto-check passed
  • Reagami

    PEZ/epupp

    Reagami UI library patterns for Epupp Squint components. An agent skill from PEZ/epupp.

    116 GitHub stars~1.4k tokensUpdated 13 days ago
    Auto-check passed
  • Epupp Design

    PEZ/epupp

    Epupp design context, principles, and visual foundation. An agent skill from PEZ/epupp.

    116 GitHub stars~610 tokensUpdated 13 days ago
    Auto-check passed
  • Babashka

    PEZ/epupp

    Babashka scripting conventions for the Epupp project. An agent skill from PEZ/epupp.

    116 GitHub stars~360 tokensUpdated 13 days ago
    Auto-check passed

Questions about Epupp Dev

What does Epupp Dev do?

Epupp architecture, development patterns, and system contracts. Epupp Dev is an agent skill from PEZ/epupp. Epupp architecture, development patterns, and system contracts.

When should I use Epupp Dev?

Epupp Dev fits situations like: modifying Epupp source code; implementing features; debugging extension behavior; working with Uniflow actions.

How do I install Epupp Dev in Claude Code?

Run `npx skills add PEZ/epupp --skill epupp-dev -a claude-code`. Or copy the skill folder (.github/skills/epupp-dev in PEZ/epupp) into .claude/skills/epupp-dev in your project. Claude Code loads it when a task matches its description.

How do I install Epupp Dev in Codex?

Run `npx skills add PEZ/epupp --skill epupp-dev -a codex`. Or copy the skill folder (.github/skills/epupp-dev in PEZ/epupp) into .agents/skills/epupp-dev in your project. Codex loads it when a task matches its description.

Can I use Epupp Dev in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add PEZ/epupp --skill epupp-dev -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/epupp-dev, .gemini/skills/epupp-dev, .github/skills/epupp-dev and .opencode/skills/epupp-dev in your project.

What does Epupp Dev need to run?

SKILL.md names no scripts, command-line tools or credentials: Epupp Dev is instructions for the agent only. Our summary lists: Python 3.

Does Epupp Dev access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Epupp Dev safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Epupp Dev use?

Epupp Dev is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Epupp Dev use?

About 2.4k tokens (SKILL.md is roughly 9.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 5.4k tokens, read only when the agent opens those files.

What are the alternatives to Epupp Dev?

Skills that share tags, products or a category with Epupp Dev: MAUI UI Test Shard Rebalancer (dotnet/maui, 23k stars), Automating Devops (telagod/code-abyss, 243 stars), PR Babysitter (openinterpreter/openinterpreter, 69k stars) and Robotics Testing (arpitg1304/robotics-agent-skills, 368 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Epupp Dev?

PEZ (a GitHub user) maintains it in PEZ/epupp, which has 116 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on September 25, 2026.

Source: PEZ/epupp on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.