magnus919_agent-skills/CHANGELOG.md

Changelog

0.7.0 (2026-08-22)

Features

• add AI operating economics skill (#368) (94b7231)
• add ascii-city-engine skill (#321) (b6a96eb)
• add data-cleaning skill bundle (#324) (dde5fa1)
• add dedicated AEO skill (b57fe51)
• add digital-twin skill (#326) (fe71295)
• Add droid-review.yml workflow (4a4d46f)
• Add droid.yml workflow (b3a1839)
• add enterprise architecture methodology (#359) (8174e82)
• add GxP AI governance coverage (#348) (f73162d)
• add multi-tenant SaaS architecture (#360) (11381f8)
• add personalized travel-guide skill (c446858)
• add software architecture methodology (#358) (73a4250)
• adr: deepen fitness-function evidence (#356) (a56360b)
• ai-governance: add ai-lifecycle-governance reference (a78adef)
• ai-governance: add fairness-bias-accountability reference (37c0b00)
• ai-governance: add foundations-and-principles reference (1e9aea7)
• ai-governance: add governance charter and use-case intake templates (452c061)
• ai-governance: add governance-maturity self-assessment scorer (3060606)
• ai-governance: add governance-operating-model reference (0d7c806)
• ai-governance: add llm-and-agent-security reference (f8b5475)
• ai-governance: add model risk assessment and model card templates (3a870a1)
• ai-governance: add privacy-and-data-governance reference (9b63dc4)
• ai-governance: add procurement, third-party, and board oversight reference (315fe93)
• ai-governance: add reference source-index for provenance (8520ce1)
• ai-governance: add regulatory-landscape reference (485c28e)
• ai-governance: add risk-management-and-frameworks reference (777c065)
• ai-governance: add third-party due diligence and board report templates (912dad6)
• ai-governance: add transparency-explainability reference (ad6b87a)
• ai-governance: add use-case risk-tier classifier script (e750ffc)
• ai-governance: assemble SKILL.md router, README, evals, and repo integration (96721d6)
• api: add landscape governance and topology (#354) (cc0274f)
• architecture: deepen codebase analysis (#355) (db2ad85)
• ascii-city-engine: deep enrichment — street furniture, signage, dense Raleigh (#322) (4480938)
• backend: add event and coexistence patterns (#361) (79caa0b)
• capacity: add tenant demand and unit cost (#363) (05e6a03)
• data-architect: add mesh and event product guidance (#350) (3d8ad44)
• diagrams: add communication quality gates (#364) (f6dc2ce)
• enrich site reliability engineering skill from O'Reilly SRE books (2c98cff)
• evals: add eval manifests for the 10 sub-skills (0096aca)
• flatten: flatten bundles/ directory into repo root (413bcb9)
• flatten: move bundle dirs to repo root (c0c7690)
• flatten: repoint catalog generators to flat skill glob (7b6f26f)
• flatten: repoint lifecycle capability matrix tooling to flat layout (42447b2)
• flatten: repoint validate-bundles glob and rebuild manifest test fixtures (b623635)
• flatten: rewrite moved-file paths for flat layout (a315b77)
• linear: project mutations and richer issue verbs in CLI (#288) (b8a5092), closes #287
• migration: add service extraction guidance (#353) (fe9b268)
• neckbeard: position-assessment entry mode and tracker-neutral intake (1c6982f)
• neckbeard: position-assessment entry mode and tracker-neutral intake (2db3bd0), closes #371
• radar: make architecture governance proportional (#357) (cc70502)
• security: deepen tenant boundary controls (#362) (016be9c)
• semantic-spacetime: add format-compliant M1 skill core (2c78dfc)
• semantic-spacetime: add M2 application references (0430e66)
• semantic-spacetime: add M4 cross-links and final validation fixes (61b3890)
• semantic-spacetime: add research-grounded semantic spacetime skill (da17e50)
• semantic-spacetime: add stdlib-only CLI and black-box test suite (2656704)
• skill: add actuarial risk modeling methodology (#332) (c990f05)
• skill: add anydoc — office documents to GitHub-Flavored Markdown (#295) (f37dc73)
• skill: add collaboration & business-app tool layer (Slack, Notion, email, CRM, payments) (#269) (3256a87)
• skill: add documents family skill (PDF / Word / Excel / PowerPoint) (#262) (c49666e)
• skill: add dsm5 — evidence-based companion to the DSM-5-TR (#276) (d346970)
• skill: add genius-life creativity practice skill (#299) (9a42be5)
• skill: add LiteLLM AI gateway operations skill (fb8f5c8)
• skill: add LiteLLM AI gateway operations skill (030e6df)
• skill: add mobile development skill (iOS / Android / Flutter / React Native) (#248) (#268) (f687218)
• skill: add Playwright skill (E2E testing + scraping + headless browsing) (#264) (ac1beb1)
• skill: add PostgreSQL operational skill (#245) (#265) (cd14da2)
• skill: add scripts, templates, and evals to backend-engineering and frontend-engineering (#256) (48c1a1e)
• skill: add telemetry skill (Prometheus + OpenTelemetry + Loki) (#246) (#266) (f83d48b)
• skill: add Terraform operational skill (#263) (00abbf9)
• skill: add vLLM inference-serving skill (#247) (#267) (6181f17)
• skill: add writing skill for the full writing lifecycle (#301) (ecdfc65)
• skill: beef up ml-engineering with scripts/templates/evals (#257) (92299e1)
• skill: cross-pollinate the new tool wave into catalog routing (#275) (6f67a34)
• skill: incorporate supabase/evals harness into supabase skill (7b5845d)
• skill: incorporate supabase/evals harness into supabase skill (4a5f18e), closes #271
• skill: promise theory — expert methodology for hybrid human+AI agent coordination (0413a05)
• skill: promise-theory catalog integration (ffd05b2)
• skill: promise-theory evals manifest + trigger probes (a30274f)
• skill: promise-theory quick start documents --json and --dry-run (31a2fd1)
• skill: promise-theory references — coordination, patterns, trust & verification (93e0d96)
• skill: promise-theory references — diagnosis & debugging + glossary (e856ddb)
• skill: promise-theory references foundations + applications-infrastructure (d01d7a1)
• skill: promise-theory scaffold — SKILL.md router + MIT LICENSE (236d678)
• skill: promise-theory script + unit tests (a2f27c0)
• skill: promise-theory templates + human README (4a780c4)
• skill: rename writing skill to writers-helper (#302) (6fe5aed)
• skills: expose audio/video recording URLs in fireflies transcripts get (#308) (#309) (b87d18e)
• skill: thicken platform-engineering with evals, templates, and deeper cloud/IaC references (#255) (abe1ab3), closes #238
• skill: travel-guide section-end footers — field notes, next-up, ghost mark (4b9d347)
• skill: travel-guide section-end footers — field notes, next-up, ghost mark (3cd705f)
• skill: travel-guide visual system — journey line, day strip, meters, photo grade (2823d82)
• skill: travel-guide visual system — journey line, day strip, meters, photo grade (3f7c46e)
• validation: enforce 60K-char cap on skill reference files (#279) (3eb7bd4)

Bug Fixes

• ai-governance: make reference cross-links resolve for the router gate (fffc38e)
• ai-governance: reconcile AI Act timeline, exit-code contract, and UK institute naming (bb25787)
• ai-governance: rephrase passages to clear the 8-word n-gram copyright gate (888872c)
• anydoc: clarify rendered-layout inspection route (#297) (43fd4b9)
• architecture: route distributed mechanisms to DDIA (#367) (00e611a)
• calculator: honest burn-multiple and runway labels, surface model assumptions (#274) (f9db3db)
• ci: align skill test discovery with coverage guardrail (4255d44)
• ci: keep epub test edit out to respect eval-coverage ratchet (84edaf7)
• ci: pin epub test deps and surface epublib skip (0dc4b6c)
• correct vllm models-check test name and stripe cancel boolean (#270) (0ff3734)
• docs: correct depth-parity comment in example bundle manifest (409826a)
• evals: drop bundles/ fixture path from eval-coverage test (cbe2772)
• evals: re-point lifecycle corpus tooling at flat layout (b7aa174)
• evals: reword expectations prose in agent-skills eval manifest (#237) (#261) (d68c1b3)
• flatten: point catalog validation and links at flat layout (df71336)
• neckbeard: address droid-review findings on position-assessment PR (074dac5)
• neckbeard: authenticate mid-flight evidence before crediting gates (c5da020)
• neckbeard: resolve all findings from rounds 3 and 4 (9bb3654)
• neckbeard: resolve round-3 droid-review consistency findings (a6ce39f)
• neckbeard: resolve round-5 findings; loop escalated as non-converging (04a9b60)
• optimize actuarial risk modeling navigation (#333) (639fdf6)
• optimize digital-twin through three SkillOpt epochs (#327) (ed68466)
• prose: refresh top-level skill path references for flat layout (3480b7f)
• raleigh: skip token-gated imagery folders in discovery and canary (#280) (e99c6d2)
• register BYOK DeepSeek model in ~/.factory/settings.json (0eaf499)
• register BYOK DeepSeek model in ~/.factory/settings.json (b18751c)
• repair dead cross-skill routing references in methodology skills (#251) (0223e2b), closes #234
• semantic-spacetime: enforce strict sst-model-v1 schema in model lint (3fc2f87)
• site-reliability: make recovery closure gate explicit (#370) (f7d550b)
• skill: guard churn prints in saas-metrics human-readable output (#259) (a45952d)
• SkillOpt 3-epoch optimization of forward-deployed-engineering bundle (#294) (cff17c5)
• SkillOpt optimization of travel-guide skill (3 epochs) (29a367e)
• SkillOpt optimize AI operating economics (#369) (1291f95)
• skill: SkillOpt 3-epoch optimization — dsm5 navigation, decisions, and answer patterns (#281) (5d10100)
• skill: split oversized dsm5 references into index + parts (#278) (c312b36)
• skills: reconcile fireflies CLI with live GraphQL schema (#290) (eadb82e)
• skills: rework moved-skill descriptions to pass quality gate (0a7c6b9)
• test: correct duplicate-catalog name fixtures for case-sensitive fs (ed3d09f)

0.6.0 (2026-08-03)

Features

• add CNCF Landscape technology selection skill (622f341)
• add comic chat PNG renderer skill (2494a66)
• add lightweight test-hardening path to neckbeard (#230) (36d3fa8)
• agent-production-operations: add agent production operations bundle (#229) (ac1ad5d), closes #201
• binary-analysis skill with Ghidra backend (0d4a365)
• binary-analysis: implement CLI skeleton with argparse, JSON envelope, and exit codes (e9397bb)
• bundles: define bundle manifests and lifecycle capability matrix (#203) (#231) (aa893e3)
• capacity-and-cost-engineering: add capacity-and-cost-engineering skill (#199) (#224) (a471888)
• ci: add linting, formatting, coverage, and security configs (476d7e1)
• ci: add mypy, radon, and deptry to CI pipeline (bfe05ef)
• conditional-customer-success: add conditional customer success skill (#192) (#218) (1dd564e)
• decompile: implement decompile command with pseudocode, address map, and diagnostics (3e4f07a)
• doctor-bootstrap: implement doctor, bootstrap, and version commands (55c1451)
• domain-foundation: implement canonical domain model with entities, enums, schemas, errors, and selectors (3481663)
• fake-backend: implement FakeAdapter with controllable backend testing (6184673)
• fix remaining agent readiness signals for level 5 (865a1e9)
• fix-coverage-config: add .coveragerc with placeholder module omissions and exclude plain-text output helpers (86b248e)
• fix-fake-adapter-env: add BINARY_FAKE_* env var support to FakeAdapter (c408259)
• fix-missing-tests: add register_binary(), clamp warnings in JSON envelope, and new tests (50e48de)
• function-queries: implement functions, disassemble, and bytes CLI commands (3cf347b)
• ghidra-adapter: implement GhidraAdapter skeleton with PyGhidra bridge (9e8234f)
• implementation-planning: add implementation planning skill (#217) (f82515e), closes #186
• import-analyze: implement binary import, analyze, and metadata commands (f4e254a)
• improve agent readiness with dev tooling, CI checks, and tests (6b44d6f)
• incident-learning: add incident-learning skill (#226) (4f14ce3), closes #200
• json-contracts: implement JSON contract consistency across all commands (801a5be)
• migration-engineering: add migration-engineering skill (#222) (652521a)
• neckbeard: issue-to-PR delivery workflow (#181-#185) (#208) (4a4958b)
• privacy-engineering: add privacy-engineering skill (#202) (#225) (6f429a9)
• product-adoption: add product adoption skill (#214) (11a6595)
• product-analytics-and-measurement: add product analytics and measurement skill (#213) (6b650b7)
• product-experimentation: add product experimentation skill (#212) (0b32a96)
• product-lifecycle-learning: add product lifecycle learning skill (#194) (#219) (2c247a1)
• product-lifecycle: add thin product-lifecycle bundle (#227) (46b92aa), closes #187
• product-operations-and-governance: add product operations and governance skill (#216) (7756266), closes #193
• product-roadmapping-and-portfolio: add product roadmapping and portfolio skill (#215) (5d95dd6), closes #189
• production-excellence: add thin production-excellence bundle (#228) (8c05a07)
• production-readiness: add production-readiness skill (#221) (2183c22)
• project-lifecycle: implement all project lifecycle commands and state machine (756b50e)
• project-system: implement workspace, manifest, locking, and cache (defe46c)
• qa-methodology: add 2 stdlib-only CLI scripts and colocated pytest suite (974fde0)
• qa-methodology: add 3 new traditional QA references (ff69b09)
• qa-methodology: add 5 fillable templates and 3 assets (3325787)
• qa-methodology: add modern agentic references (ai-code-quality-gates, agentic-eval-design) (aa8c027)
• qa-methodology: add modern QA career levels and SDET engineering references (9b10ff5)
• qa-methodology: add mutation-guided test hardening evidence workflow (8bd042e), closes #209
• qa-methodology: add schema-v1 evals with 7 output-quality cases (107d6d1)
• qa-methodology: rebuild as thin-index QA/SDET skill bundle (726de89)
• qa-methodology: rebuild SKILL.md as thin index and rewrite README.md (8b0220b)
• qa-methodology: refresh 5 traditional QA support references (5bad13a)
• qa-methodology: refresh traditional QA core references and split monolithic file (966edcf)
• reference-queries: implement xrefs, callers, callees, and callgraph CLI commands (93b0431)
• release-engineering: add release engineering skill bundle (818d8ca)
• reporting: implement export-report and audit commands (010a308)
• resilience-and-recovery: add resilience-and-recovery skill (#223) (c032580)
• safety-hardening: implement path security, output limits, memory limits, and report containment (e62cf47)
• search-trace: implement search and trace commands with cross-area integration flows (1ead735)
• security-rules: implement suspicious-apis and capability-map commands (d005b9a)
• skill-content: write SKILL.md and README.md for binary-analysis skill (72d8462)
• skill-evals: write 6 eval cases for binary-analysis skill (f79124d)
• skill-references: write 11 reference files for binary-analysis skill (a799e0f)
• structural-queries: implement sections, entrypoints, imports, exports, symbols, and strings CLI commands (7df699b)
• triage-diagnostics: implement triage and diagnostics commands (0e05960)
• worker: implement optional local worker with start/stop/status commands and one-shot fallback (5e8f1d3)

Bug Fixes

• ci: add binary-analysis to deptry extend_exclude (8ea4053)
• ci: add sys.path setup in test __init__.py for unittest discover (4c7f5c4)
• ci: add sys.path setup to all test sub-package __init__.py files (f9354df)
• ci: add sys.path setup to individual binary-analysis test files (736859e)
• ci: prevent sys.modules cross-contamination between test directories in check-artifacts (96f0bd4)
• ci: use full path for deptry binary-analysis exclude (b56183b)
• permit removal of eval-covered skills (687d7d7)
• product-strategy: correct stale RICE reference (#211) (8d30f22)
• release-engineering: address review findings in DORA asset, cadence, link, metrics script (2b90656)
• release-engineering: align 0.x version bumps with Release Please's minor-feature policy
• release-engineering: validate Keep a Changelog and Release Please formats
• relocate binary analysis skill (6387f7f)
• require dated commitments for CoS tasks (857dfc6)
• security-ship: ensure diagnostics have both recoverable values and add pagination cursors to security commands (d04af37)
• security-ship: fix --limit routing, truncation warnings, and clean confirmation order (7b0109b)
• security-ship: fix three scrutiny blocking issues (a20adde)
• state CoS task-capture boundary (f285316)

Reverts

• restore clean base64 import in helpers.py (746df95)

0.5.0 (2026-07-27)

Features

• add evidence-aware life coaching skill (#146) (795d142)
• add Gibson cyberpunk skill (#149) (daece12)
• add harness adapter contract and reproducible eval run artifacts (#130) (e83558a)
• add llama.cpp operations skill (#145) (58457b4)
• add release-grade repeated and rubric-based skill evaluation (#136) (0a80bda), closes #106
• expand Raleigh civic data skill (eff14ca)
• grafana: add operations skill (4c824b7)
• grafana: add operations skill (e304366)
• raleigh: add fire protection proximity lookup (#144) (1f5cd6a)
• raleigh: add first-class RFD incident workflows and normalize the 2026 schema (#140) (0ca5c24), closes #122
• raleigh: add first-class RPD incident workflows (#139) (50be18b), closes #121
• raleigh: add guarded fire report lookups (12b650e)
• raleigh: add guarded fire report lookups (4eaf5aa)
• raleigh: add official public safety statistics (b030f26)
• raleigh: add Raleigh-Wake ECC active incident feed adapter (#142) (d4ab656)
• raleigh: expand civic data skill (903ea05)
• raleigh: expose official public safety statistics (dcd6f59)
• run isolated paired candidate and baseline skill evaluations (#133) (390f3e3)
• skill: add PACE plan lifecycle (e62152c)
• skill: add PACE plan lifecycle (64383d9)
• validate eval manifest coverage states (#129) (a617cca)

Bug Fixes

• align eval artifact schemas with runtime (f49e670)
• allow incidental text in cyberpunk image briefs (6b467df)
• complete eval ratchet verification (#109) (1e10a56)
• enforce eval coverage ratchet for complete changed skill directories (#107) (fc8a895), closes #102
• harden paired evaluation isolation (cb1579f)
• life-coach: simplify capability onboarding (#148) (fdb00b4)
• llama-cpp: harden operational diagnostics (#147) (86ac280)
• raleigh: bound publication source contracts (cc26b4c)
• raleigh: enforce publication request contracts (1bd7b06)
• raleigh: fail on missing statistics sections (8becc22)
• raleigh: normalize report probe failures (f3dfb68)
• raleigh: normalize statistics source failures (07c2ae2)
• raleigh: reject malformed content identifiers (5f74b4d)
• raleigh: restore live RPD queries (cfbcfb0)
• raleigh: restore live RPD queries (4a23562)
• raleigh: validate included resource identifiers (9f2e4b6)
• raleigh: validate statistics response shapes (bf69aa1)
• reject duplicate eval fixture paths (cbe1b69)
• require causal layer maps for cyberpunk image briefs (34fab96)
• require causal layer maps for cyberpunk image briefs (34fab96)
• require causal layer maps for cyberpunk image briefs (c514c85)
• require evidence for Grafana dashboard defaults (4b60a2e)
• require evidence for Grafana dashboard defaults (4ecc536)
• skill: finalize PACE delivery evidence (f62a5de)
• skills: rename gibson-cyberpunk to cyberpunk (#150) (1005462)
• verification: preserve requested source fidelity (#127) (547aed2)

0.4.0 (2026-07-22)

Features

• add Claude Code plugin marketplace (metadata-only catalog) (#80) (3c8b6cd), closes #76
• add Codex plugin packaging (single-plugin, metadata-only) (#83) (0f2aaf1)
• add generated llms.txt skill index (#93) (cde4a67), closes #78
• add neckbeard, an evidence-driven SDLC skill bundle (#81) (e167718)
• add programming-principles skill (14 classic software books) (#92) (1037324)
• eval coverage ratchet — gate new skills, track coverage, ratchet thresholds (#99) (be0c8df), closes #90
• qa-methodology: add ci-failure-triage and test-debugging references (#94) (08fafac)
• validate changed skill descriptions (#97) (c092a14)

Bug Fixes

• close skill quality lint gaps (#98) (dd457f1)
• correct bundle skill paths in Codex plugin manifest (#91) (3d650da)
• pin release-please manifest to v0.3.0 (#72) (0e8bc26)
• SkillOpt 3-epoch optimization of neckbeard — description, routing, stage alignment (#84) (051bf2a)