@zeldrisho/pi-anthropics-skills 0.1.1 → 0.1.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/CHANGELOG.md CHANGED
@@ -7,6 +7,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
7
7
 
8
8
  ## [Unreleased]
9
9
 
10
+ ## [0.1.2] - 2026-09-28
11
+
12
+ ### Changed
13
+
14
+ - Restored pinned upstream skill content, removed bundled auxiliary files, and corrected package license metadata.
15
+ - Clarified documentation API guidance and required evidence for technical-debt findings; removed unrelated text appended to the license.
16
+
10
17
  ## [0.1.1] - 2026-09-23
11
18
 
12
19
  ### Changed
@@ -20,6 +27,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
20
27
 
21
28
  - `documentation`, `testing-strategy`, and `tech-debt` Pi skills from Anthropic's knowledge-work plugins
22
29
 
23
- [Unreleased]: https://github.com/zeldrisho/pi-packages/compare/pi-anthropics-skills-v0.1.1...HEAD
30
+ [Unreleased]: https://github.com/zeldrisho/pi-packages/compare/pi-anthropics-skills-v0.1.2...HEAD
31
+ [0.1.2]: https://github.com/zeldrisho/pi-packages/compare/pi-anthropics-skills-v0.1.1...pi-anthropics-skills-v0.1.2
24
32
  [0.1.1]: https://github.com/zeldrisho/pi-packages/compare/pi-anthropics-skills-v0.1.0...pi-anthropics-skills-v0.1.1
25
33
  [0.1.0]: https://github.com/zeldrisho/pi-packages/releases/tag/pi-anthropics-skills-v0.1.0
package/LICENSE CHANGED
@@ -1,6 +1,3 @@
1
- Anthropic engineering skill adaptations are licensed under Apache-2.0. Package-authored evaluation cases are offered under MIT. Source: https://github.com/anthropics/knowledge-work-plugins/tree/93d82a54e9516d172c1d0a010c1c09d56bf9d11b/engineering/skills
2
-
3
-
4
1
  Apache License
5
2
  Version 2.0, January 2004
6
3
  http://www.apache.org/licenses/
package/README.md CHANGED
@@ -1,6 +1,12 @@
1
1
  # @zeldrisho/pi-anthropics-skills
2
2
 
3
- Pi-portable adaptations of Anthropic's [`engineering` skills](https://github.com/anthropics/knowledge-work-plugins/tree/93d82a54e9516d172c1d0a010c1c09d56bf9d11b/engineering/skills): `documentation`, `testing-strategy`, and `tech-debt`. These files have since been substantially revised for Agent Skills and Pi workflows; they are not verbatim upstream copies and are not endorsed by Anthropic.
3
+ Pi package of selected [Anthropic engineering skills](https://github.com/anthropics/knowledge-work-plugins). This package is not endorsed by Anthropic.
4
+
5
+ | Skill | Purpose |
6
+ | ------------------ | -------------------------------------------- |
7
+ | `documentation` | Write and maintain engineering documentation |
8
+ | `testing-strategy` | Plan risk-based testing |
9
+ | `tech-debt` | Audit and prioritize technical debt |
4
10
 
5
11
  ## Install
6
12
 
@@ -10,8 +16,8 @@ pi install npm:@zeldrisho/pi-anthropics-skills
10
16
  pi install -l npm:@zeldrisho/pi-anthropics-skills
11
17
  ```
12
18
 
13
- Select individual skills with [Pi package filtering](../../docs/package-filtering.md). Upstream provenance and license scope are documented in [`LICENSE`](LICENSE).
19
+ Select individual skills with [Pi package filtering](../../docs/package-filtering.md).
14
20
 
15
21
  ## License
16
22
 
17
- Adapted upstream material is under Apache-2.0; package-authored material is under MIT where separable. See [`LICENSE`](LICENSE) for scope and attribution.
23
+ The upstream skills are Apache-2.0. See [`LICENSE`](LICENSE) for terms and attribution.
package/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "@zeldrisho/pi-anthropics-skills",
3
- "version": "0.1.1",
4
- "description": "Pi adaptations of Anthropic engineering skills, revised for Agent Skills workflows",
3
+ "version": "0.1.2",
4
+ "description": "Anthropic engineering skills packaged for Pi",
5
5
  "keywords": [
6
6
  "documentation",
7
7
  "pi",
@@ -14,7 +14,7 @@
14
14
  "bugs": {
15
15
  "url": "https://github.com/zeldrisho/pi-packages/issues"
16
16
  },
17
- "license": "(Apache-2.0 AND MIT)",
17
+ "license": "Apache-2.0",
18
18
  "author": "Zeldris",
19
19
  "repository": {
20
20
  "type": "git",
@@ -1,27 +1,54 @@
1
1
  ---
2
2
  name: documentation
3
- description: Use this skill when writing or maintaining engineering documentation such as READMEs, API references, runbooks, architecture documents, or onboarding guides for a specific audience.
3
+ description: Write and maintain technical documentation. Trigger with "write docs for", "document this", "create a README", "write a runbook", "onboarding guide", or when the user needs help with any form of technical writing — API docs, architecture docs, or operational runbooks.
4
4
  ---
5
5
 
6
6
  # Technical Documentation
7
7
 
8
- Create documentation that helps its intended reader complete a task accurately. Keep procedures and claims grounded in the project, not assumptions.
8
+ Write clear, maintainable technical documentation for different audiences and purposes.
9
9
 
10
- ## Workflow
10
+ ## Document Types
11
11
 
12
- 1. Identify the reader, their goal, and the document type. For a README, optimize for first successful use; for a runbook, incident recovery; for API docs, correct integration; for architecture docs, decisions and system boundaries; for onboarding, safe completion of common tasks.
13
- 2. Inspect the implementation and authoritative sources (configuration, API schemas, existing docs, tests, CI). Reuse verified commands and names; mark unresolved behavior as a question rather than guessing.
14
- 3. Lead with the information needed to act. Order steps by dependency and include prerequisites, expected results, and failure or rollback guidance when relevant.
15
- 4. Include a minimal working example for interfaces or procedures. Verify syntax, paths, options, outputs, and links against available sources; do not invent API behavior.
16
- 5. Link to the source of truth instead of duplicating policy or reference material. Keep duplicated facts minimal and clearly scoped.
17
- 6. Review for the intended reader: can they find the next action, understand terminology, and distinguish required from optional steps? Remove stale, redundant, or generic text.
12
+ ### README
18
13
 
19
- ## Useful content by document type
14
+ - What this is and why it exists
15
+ - Quick start (< 5 minutes to first success)
16
+ - Configuration and usage
17
+ - Contributing guide
20
18
 
21
- - **README:** purpose, prerequisites, quick start, configuration, common usage, and links to contribution/support guidance.
22
- - **API reference:** authentication, request/response shapes, errors, pagination/limits where applicable, and runnable examples.
23
- - **Runbook:** trigger/conditions, access prerequisites, ordered procedure, verification, rollback, and escalation when known.
24
- - **Architecture:** context, goals, components/data flow, key decisions, trade-offs, and boundaries.
25
- - **Onboarding:** environment setup, system relationships, common tasks, and verified help channels.
19
+ ### API Documentation
26
20
 
27
- Adapt these lists; do not add sections that do not help the reader.
21
+ - Endpoint reference with request/response examples
22
+ - Authentication and error codes
23
+ - Rate limits and pagination, when supported by the API
24
+ - SDK examples, when available and verified
25
+
26
+ ### Runbook
27
+
28
+ - When to use this runbook
29
+ - Prerequisites and access needed
30
+ - Step-by-step procedure
31
+ - Rollback steps
32
+ - Escalation path
33
+
34
+ ### Architecture Doc
35
+
36
+ - Context and goals
37
+ - High-level design with diagrams
38
+ - Key decisions and trade-offs
39
+ - Data flow and integration points
40
+
41
+ ### Onboarding Guide
42
+
43
+ - Environment setup
44
+ - Key systems and how they connect
45
+ - Common tasks with walkthroughs
46
+ - Who to ask for what
47
+
48
+ ## Principles
49
+
50
+ 1. **Write for the reader** — Who is reading this and what do they need?
51
+ 2. **Start with the most useful information** — Don't bury the lede
52
+ 3. **Show, don't tell** — Code examples, commands, screenshots
53
+ 4. **Keep it current** — Outdated docs are worse than no docs
54
+ 5. **Link, don't duplicate** — Reference other docs instead of copying
@@ -1,39 +1,33 @@
1
1
  ---
2
2
  name: tech-debt
3
- description: Use this skill when auditing technical debt or code health, deciding whether to refactor, or prioritizing a maintainability backlog.
3
+ description: Identify, categorize, and prioritize technical debt. Trigger with "tech debt", "technical debt audit", "what should we refactor", "code health", or when the user asks about code quality, refactoring priorities, or maintenance backlog.
4
4
  ---
5
5
 
6
6
  # Tech Debt Management
7
7
 
8
- Build an evidence-based backlog of maintainability risks and opportunities; do not label unfamiliar or merely disliked code as debt without explaining its cost.
8
+ Systematically identify, categorize, and prioritize technical debt.
9
9
 
10
- ## Workflow
11
-
12
- 1. Clarify the audit scope, goals, constraints, and time horizon. Inspect relevant code, tests, dependency/configuration files, operational docs, and recent history or incidents when available.
13
- 2. Record each candidate with a specific location or artifact, observed condition, current/potential consequence, and evidence. Separate confirmed problems from hypotheses and identify missing evidence.
14
- 3. Categorize candidates: code, architecture, test, dependency, documentation, or infrastructure. Describe overlap once and link related items rather than double-counting.
15
- 4. Estimate impact, risk, and effort using the anchors below. Explain unusual scores and note dependencies or uncertainty; do not imply numeric precision.
16
- 5. Prioritize changes that reduce material risk or recurring friction. Suggest a bounded first step, success/exit criteria, and a suitable phase that can fit alongside feature work.
10
+ ## Categories
17
11
 
18
- ## Relative scoring (1–5)
12
+ | Type | Examples | Risk |
13
+ | ----------------------- | ---------------------------------------------------- | ------------------------ |
14
+ | **Code debt** | Duplicated logic, poor abstractions, magic numbers | Bugs, slow development |
15
+ | **Architecture debt** | Monolith that should be split, wrong data store | Scaling limits |
16
+ | **Test debt** | Low coverage, flaky tests, missing integration tests | Regressions ship |
17
+ | **Dependency debt** | Outdated libraries, unmaintained dependencies | Security vulns |
18
+ | **Documentation debt** | Missing runbooks, outdated READMEs, tribal knowledge | Onboarding pain |
19
+ | **Infrastructure debt** | Manual deploys, no monitoring, no IaC | Incidents, slow recovery |
19
20
 
20
- - **Impact:** 1 = localized inconvenience; 3 = recurring team or user cost; 5 = major reliability, security, delivery, or business impact.
21
- - **Risk:** 1 = unlikely/low consequence; 3 = plausible regression, outage, or maintenance failure; 5 = credible severe or broad harm.
22
- - **Effort:** 1 = small, isolated change; 3 = several files or coordinated tests; 5 = broad redesign/migration or substantial unknowns.
21
+ ## Prioritization Framework
23
22
 
24
- Optional ranking score: `(impact + risk) × (6 − effort)`. Use it as a discussion aid, not an objective measure; urgency, dependencies, and confidence can override the ranking.
23
+ Score each item on:
25
24
 
26
- ## Categories
25
+ - **Impact**: How much does it slow the team down? (1-5)
26
+ - **Risk**: What happens if we don't fix it? (1-5)
27
+ - **Effort**: How hard is the fix? (1-5, inverted — lower effort = higher priority)
27
28
 
28
- | Type | Examples to investigate | Typical consequence |
29
- | -------------- | ------------------------------------------------- | --------------------------------- |
30
- | Code | Duplication, brittle coupling, unclear invariants | Defects, slow changes |
31
- | Architecture | Misplaced boundaries, unsafe data flows | Scaling or change constraints |
32
- | Test | Missing critical-path checks, flaky suites | Regressions, slow feedback |
33
- | Dependency | Unsupported or risky dependency versions | Security, compatibility |
34
- | Documentation | Stale procedures or undocumented constraints | Onboarding and operational errors |
35
- | Infrastructure | Manual recovery/deployments, missing safeguards | Incidents and recovery time |
29
+ Priority = (Impact + Risk) x (6 - Effort)
36
30
 
37
31
  ## Output
38
32
 
39
- Present a short summary, then prioritized items with evidence/location, category, consequence, impact/risk/effort and confidence, recommended next step, dependencies, and completion criteria. Group into practical phases. Avoid unsupported claims about coverage, vulnerabilities, or dependency status; state what needs verification.
33
+ Produce a prioritized list with evidence and source locations, observed consequences, estimated effort, business justification for each item, and a phased remediation plan that can be done alongside feature work.
@@ -1,31 +1,33 @@
1
1
  ---
2
2
  name: testing-strategy
3
- description: Use this skill when planning tests, deciding what tests to write, evaluating existing test coverage, or choosing test architecture for a software change.
3
+ description: Design test strategies and test plans. Trigger with "how should we test", "test strategy for", "write tests for", "test plan", "what tests do we need", or when the user needs help with testing approaches, coverage, or test architecture.
4
4
  ---
5
5
 
6
6
  # Testing Strategy
7
7
 
8
- Produce a risk-based plan grounded in the codebase and the team's ability to run and maintain tests.
8
+ Design effective testing strategies balancing coverage, speed, and maintenance.
9
9
 
10
- ## Workflow
10
+ ## Testing Pyramid
11
11
 
12
- 1. Clarify the change, user-visible behavior, critical data, and constraints (runtime, CI time, test environment).
13
- 2. Inspect implementation, existing tests, test commands/configuration, and recent relevant failures if available. Identify what is already covered before proposing additions.
14
- 3. Trace critical paths and trust boundaries. Include expected behavior, invalid inputs, failure paths, and state/data integrity.
15
- 4. Choose the smallest test level that gives meaningful confidence: unit for isolated logic, integration for component boundaries, end-to-end for essential user journeys. Add contract tests where independently deployed consumers depend on an interface.
16
- 5. Prioritize cases by impact and likelihood; note test data/setup, dependencies, and whether each case is automated, manual, or currently blocked.
17
- 6. Recommend coverage targets only when tied to a risk or meaningful behavior; do not prescribe a universal percentage.
12
+ ```
13
+ / E2E \ Few, slow, high confidence
14
+ / Integration \ Some, medium speed
15
+ / Unit Tests \ Many, fast, focused
16
+ ```
18
17
 
19
- ## Coverage considerations
18
+ ## Strategy by Component Type
20
19
 
21
- - APIs: business rules, authorization, request/response contracts, error mapping.
22
- - Data workflows: validation, transformations, idempotency, partial failure, recovery.
23
- - Frontends: key interactions, accessibility, loading/error states; use visual tests only where visual regressions matter.
24
- - Infrastructure: deployment/configuration smoke tests and resilience/load tests when justified by impact.
25
- - Scripts and migrations: include tests when they transform important data, change permissions, or can cause costly/irreversible effects. Do not exclude work just because it is a one-off.
20
+ - **API endpoints**: Unit tests for business logic, integration tests for HTTP layer, contract tests for consumers
21
+ - **Data pipelines**: Input validation, transformation correctness, idempotency tests
22
+ - **Frontend**: Component tests, interaction tests, visual regression, accessibility
23
+ - **Infrastructure**: Smoke tests, chaos engineering, load tests
26
24
 
27
- Avoid low-value tests that only restate implementation or exercise trivial framework behavior. Prefer observable assertions over implementation details.
25
+ ## What to Cover
26
+
27
+ Focus on: business-critical paths, error handling, edge cases, security boundaries, data integrity.
28
+
29
+ Skip: trivial getters/setters, framework code, one-off scripts.
28
30
 
29
31
  ## Output
30
32
 
31
- Give a prioritized plan with behavior/risk, test level, concrete cases, existing coverage/gaps, and execution constraints. Distinguish recommended tests from optional follow-up work; do not invent coverage claims or targets without repository evidence.
33
+ Produce a test plan with: what to test, test type for each area, coverage targets, and example test cases. Identify gaps in existing coverage.
@@ -1,23 +0,0 @@
1
- {
2
- "skill_name": "documentation",
3
- "evals": [
4
- {
5
- "id": 1,
6
- "prompt": "Update this README's quick start based on the package scripts and config. Check commands actually exist.",
7
- "expected_output": "A concise README section using verified commands/config, with source paths and no invented behavior.",
8
- "assertions": [
9
- "Commands are verified against project files",
10
- "Unverified behavior is not stated as fact"
11
- ]
12
- },
13
- {
14
- "id": 2,
15
- "prompt": "Write an incident runbook for the service, but its rollback procedure is unclear in the docs.",
16
- "expected_output": "A procedure grounded in known steps, with unknown rollback details clearly called out rather than fabricated.",
17
- "assertions": [
18
- "Unknown rollback details are explicit",
19
- "Procedure includes verification or escalation when known"
20
- ]
21
- }
22
- ]
23
- }
@@ -1,23 +0,0 @@
1
- {
2
- "skill_name": "tech-debt",
3
- "evals": [
4
- {
5
- "id": 1,
6
- "prompt": "Audit this package for technical debt. Give evidence and a practical first phase; don't just list style preferences.",
7
- "expected_output": "Prioritized candidates tied to artifacts, consequences, confidence, and actionable completion criteria.",
8
- "assertions": [
9
- "Each finding has evidence/location and consequence",
10
- "Scoring uncertainty and completion criteria are stated"
11
- ]
12
- },
13
- {
14
- "id": 2,
15
- "prompt": "A dependency looks old. Should we replace it? We have no incident history handy.",
16
- "expected_output": "The agent verifies support/security status before claiming risk, records uncertainty, and suggests a bounded next step.",
17
- "assertions": [
18
- "Does not claim vulnerability or unsupported status without evidence",
19
- "States what needs verification"
20
- ]
21
- }
22
- ]
23
- }
@@ -1,23 +0,0 @@
1
- {
2
- "skill_name": "testing-strategy",
3
- "evals": [
4
- {
5
- "id": 1,
6
- "prompt": "Plan tests for an endpoint that updates a user's billing address. Inspect existing tests first and cover authorization and validation.",
7
- "expected_output": "A prioritized, repository-grounded plan accounting for existing coverage and unauthorized access.",
8
- "assertions": [
9
- "Existing tests are inspected before gaps are claimed",
10
- "Unauthorized access and invalid input are covered"
11
- ]
12
- },
13
- {
14
- "id": 2,
15
- "prompt": "This one-off migration rewrites customer identifiers and can be rerun after failure. What tests should we add?",
16
- "expected_output": "Risk-based tests for correctness, idempotency, partial failure/recovery and data integrity.",
17
- "assertions": [
18
- "Does not exclude tests because migration is one-off",
19
- "Covers rerun/idempotency and data integrity"
20
- ]
21
- }
22
- ]
23
- }