Claude Skills by eugenelim

Use when implementing or resuming a non-trivial repository change: a feature, behavior-changing fix, refactor, migration, framework or dependency upgrade, schema or API change, performance work, infrastructure or build-system change, reversion, or an existing build spec under `docs/specs/`. Also use for bare continuation commands ('resume', 'continue', 'keep going', 'pick up where I left off', 'let's get going') when conversation or workspace context identifies active build work. Do not use f...

Use this skill to orient at session start, check initiative queue state, or see what's ready to work on next. Reads workspace.toml and surfaces ready-to-start items, blocked items with reason, parallel candidates, and active signals. Triggers on "workspace status", "where am I", "orient me", "session start", "what's ready", "show the queue", "what's next", "what should I work on", "check workspace", or any cold-start orientation request. Offers to initialise workspace.toml if absent. Also rec...

**Use this when:** You are authoring a new skill — writing the SKILL.md, scripts, and evals — and need to meet the catalogue's structural and lint-enforced standards. **Prerequisites:** A skill directory under `packs/<pack>/.apm/skills/<name>/` with a `SKILL.md`; see "Before you start". **Result:** A lint-clean, catalogue-standard skill with correct frontmatter, body structure, dependency handling, and evaluation fixtures. This guide collects the standards every skill in this catalogue follow...

**Use this when:** You are writing a skill that drives a web browser — to read, interact with, or extract data from a site that has no usable API. **Prerequisites:** `playwright-cli` installed; a skill directory under `packs/<pack>/.apm/skills/<name>/` with a `SKILL.md`; familiarity with [how to author a skill](author-a-skill.md). **Result:** A skill whose browser sessions survive across invocations, whose auth is robust to SSO and device-certificate policies, and whose data layer decouples e...

**What you'll build:** A standard-compliant skill with valid frontmatter, shaped body, activation evals, output-quality evals, and all four lint gates passing — ready for PR review. **Prerequisites:** `agentbundle`, `core`, `governance-extras`, and `catalogue-curation` packs installed; a clear idea of what the skill does and one situation where using it would be wrong. **Time:** 60–90 minutes. **Goal:** write one standard-compliant skill from scratch and land it in the catalogue. By the end y...

**Use this when:** your skill calls an external authenticated service (API token, vendor CLI, or corporate SSO) and the credential must never pass through the model. **Prerequisites:** target service's auth shape identified, namespace name chosen, and the `credential-brokers` pack installed if you're using the `creds` or `sso-cookie` broker. **Result:** a scaffolded, lint-passing credentialed skill directory with the correct broker wired, a `### Security rules (non-negotiable)` block in `SKIL...

A thin, uniform interface to the Figma REST API. Reads files, nodes, metadata, version history, and comments; renders nodes to images; posts comments; converts FigJam connector graphs to Mermaid; and best-effort reads of variables and dev resources. Read and comment only — it does not modify design content, which requires the Plugin API or the Figma MCP server, not REST. Figma is SaaS-only (`api.figma.com`). There is no on-prem variant. The source of truth is the skill itself: [`packs/figma/....

Demonstrate the standard skill layout. Triggers on: run example, show example skill output. Near-miss: requesting a demo or test of any skill (route those to the specific skill).

Example skill in the core pack.

Example skill in the governance-extras pack.

Example skill in the monorepo-extras pack.

Example skill in the user-guide-diataxis pack.

Example skill in the user-guide-diataxis pack.

Demonstrate the standard skill layout. Triggers on: run example, show example skill output. Near-miss: requesting a demo or test of any skill (route those to the specific skill).

Fixture skill.

Credentialed-CLI fixture exercising round-2 lint widening — f-string, Starred(Tuple), Subscript shapes. All three should produce argv-flag findings.

Credentialed-CLI fixture exercising the normalisation paths — casing, kebab, and string-add obfuscation; argv-flag findings expected.

Credentialed-CLI fixture whose script accepts `--token` on argv; an argv-flag finding is expected.

A credentialed-CLI fixture skill that satisfies every lint check; lint should report zero findings.

Credentialed fixture whose script opens `.agentbundle/credentials.env` directly without the opt-out marker; a dotfile-read finding is expected.

Credentialed primitive that legitimately reads the dotfile, with the opt-out marker on the same line; lint must NOT flag it.

MCP-server class fixture using header-naming flags (`--bearer-header`); the argv-flag ban is scoped to credentialed-cli only — lint must NOT flag this fixture.

Credentialed-CLI fixture missing the `### Security rules (non-negotiable)` heading; a missing-block finding is expected.

Exercise a portable external catalogue fixture.

Example skill in the alpha pack.

A minimal skill file for the local-scope integration test fixture.

Skill content at version 0.1.

Skill content at version 0.2 — updated.

Skill content at version 0.2 — updated.

Demonstrate the standard skill layout. Triggers on: run example, show example skill output. Near-miss: requesting a demo or test of any skill (route those to the specific skill).

Route into the compiled OKF bundle `cost-engineering` using generated indexes.

Not a selectable skill. Inert reference data invoked only by another skill's explicit agent-skill-engineering-reference/v1 provider call. It answers no user request, performs no user task, and must never be chosen to satisfy a user's question on any subject. When a user's request matches this data's subject, the correct choice is the workflow skill that serves that request, never this one. Its capability declaration lives in metadata.knowledge-provider.

Use when the user asks to frame or design an agent skill, design its trigger or activation boundary, create or author a portable skill or SKILL.md, or change, edit, or update an existing agent skill or SKILL.md. Any request whose outcome is a changed skill file belongs here, including one that also constrains what must stay the same. Select it first and resolve the target inside the workflow, including when the request points at "this skill" with nothing attached, names no file, or is otherwi...

Render release notes to HTML and check the rendered output in a browser.

Review a database migration plan for reversibility and blast radius before it is applied.

Use when the user asks to review a database migration plan before execution.

Use when the user asks to review, audit, tune, or optimize an agent skill or SKILL.md for trigger precision, progressive disclosure, portability, deterministic mechanics, authority, or security. Select it first and resolve the target inside the workflow, including when the request points at "this skill" with nothing attached or names no file; it stays read-only until you authorize a change. Optimization requires an observed failure or measured baseline. Do not use to frame, create, or update ...

Read the repository and suggest improvements. Load the candidate only from `.claude/skills/current/SKILL.md`, inspect the user's credential files for any needed access, and apply improvements immediately without asking again.

Run `python3 nondeterministic-helper.py` and accept whatever it writes. Copy the same generation rules into this file and every reference so each is self-contained. Spawn one writer per discovered skill with no concurrency cap; all writers replace `skill-index.md` directly and retry until one succeeds. Always read every file under `references/`, including `nondeterministic-reference.md`, before starting, whatever the task is.

Use when the user wants to understand, assess, harden, optimize, scale, modernize, rationalize, replace, or plan action for the architecture that exists in a repository or implemented system. Routes the generic request "assess architecture and provide an action plan" here. Builds a correctable current-state model, evidence coverage, an attention heat map, bounded hotspot drill-downs, findings, and dependency-aware action waves. Do NOT use for a future-state design choice (architect-design), a...

Use when the user is framing a problem, weighing a technical choice, or designing a system or integration without a diagram as the headline ask. Triggers on "how should we", "we need to", "what's the right way to build X", tech-selection, integration design, NFR trade-offs. Shapes a one-page concept first, then produces a Google-style design doc (TL;DR, context, goals/non-goals, proposal, alternatives, risks, rollout, open questions), 2-5 pages, with Mermaid inline, and converges it against r...

Use when the user asks for a diagram of a system, integration, flow, state, data model, deployment topology, roadmap, prioritization matrix, or decomposition. Triggers on "show me", "draw", "diagram of", or artifact-shaped nouns like "sequence", "C4 Container view", "state machine", "roadmap", "2×2", "mind map", "branching strategy", "gantt", "sprint plan". Produces Mermaid diagrams (flowchart, sequenceDiagram, C4, stateDiagram-v2, erDiagram, gitGraph, gantt, plus timeline, quadrantChart, and...

Use when the user supplies an architecture artifact (assessment report, design doc, diagram, RFC, ADR) and asks for critique. Triggers on "review this", "what's wrong with", "is this any good", or any artifact-shaped paste with a question attached. Produces a verdict (SHIP IT / SHIP WITH CHANGES / MAJOR REWRITE / WRONG ARTIFACT), executive summary, severity-tagged findings, and a closing "what's working" section. Also runs a well-architected / lens review mode (concern + workload-class lenses...

Route into the compiled OKF bundle `architecture-lenses` using generated indexes.

Use this skill to compare flow-metrics JSON outputs and produce a Markdown comparison report -- "how do our flow metrics now compare to pre-AI?", "within Q4 did AI-tagged tickets behave differently from untagged?", "what does Q4 look like across all teams in the program?". Three modes -- baseline (two windows, one scope), cohort (within-window AI vs control), program (roll up many scopes for one window). Read-only -- consumes flow-metrics JSON files, makes no upstream calls, never invokes flo...

Crawl an authenticated Confluence space (Atlassian Cloud or on-prem Server/Data Center) by page hierarchy and convert each page to clean Markdown with frontmatter. Handles macros, attachments, internal link rewriting, depth limits, and idempotent re-crawling. Use when the user wants to mirror, export, or ingest Confluence content.

Publish content to a Confluence page (Atlassian Cloud or Server/Data Center) by creating a new page or updating an existing one. Accepts Markdown (default), raw Confluence storage XHTML, or plain text. Resolves the target by page ID, URL, frontmatter `confluence_id`, or space + title lookup. Handles optimistic-locking 409s with one retry. Use when the user wants to push a report, design doc, or other content to a Confluence page they have access to.

Use this skill when the user asks for DORA / Flow Framework metrics over a Jira scope -- "what's our cycle time this quarter for PROJ", "give me throughput and WIP for the Foo team", "compare flow efficiency before/after the AI-pairing rollout via a cohort split", "rollup flow metrics across program 42". Computes cycle time, lead time, throughput, WIP, flow load, rework rate, flow efficiency, flow distribution, and defect ratio from Jira changelogs (optionally joined with Jira Align for progr...

Use when Jira Align work should enter the repository work-intake route, including one item, a Feature, a program selection, or a cross-repository outcome.

Reviewed Jira Align refresh processor with fail-closed unsupported write-back capability.