claude-mods/CHANGELOG.md

Changelog

All notable changes to claude-mods are documented here. Format follows Keep a Changelog; versions follow Semantic Versioning. Fuller narrative entries for feature releases live in the README "Recent Updates" section.

Unreleased

[3.1.0] - 2026-06-17

Added

• mapbox-ops skill - advanced Mapbox GL JS toolkit for the web (v3, not the native SDKs): custom markers, thematic dataviz, 3D/terrain, cinematic camera, style composition, expressions, performance, and hard-won gotchas across 14 reference files. Ships a headless-Playwright map verifier (screenshot_map.py - asserts a marker projects to its lng/lat) and a stdlib-only check-mapbox-facts.py §7 staleness verifier: offline asserts the v3 Standard config enums, terrain tileset IDs, and weather/camera version gates stay internally consistent; live resolves the third-party style URLs and flags a GL JS major bump past v3. 37-assertion offline self-test.
• pypi-ops skill - publish Python packages to PyPI via OIDC Trusted Publishing with PEP 740 attestations (gh-action-pypi-publish), not stored API tokens. Covers first-publish pending-publisher setup, the invalid-publisher / already-exists failure ladder, TestPyPI dry runs, release-environment approval gates, local uv publish / twine, and a stale-OIDC-federation audit (the Mini Shai-Hulud publish-token vector).
• docs/SKILL-CREATION-PROTOCOL.md - the canonical "how to build a claude-mods skill" sequencing doc (warranted? → frontmatter → body → resources → tests → repo wiring → ship). Cites rather than restates the layer-owning docs (skill-creator, SKILL-SUBAGENT-REFERENCE, naming-conventions, SKILL-RESOURCE-PROTOCOL) and carries a precedence table for when they disagree. skill-agent-updates.md now routes here first.

Changed (terminal output)

• Terminal design system promoted from experimental to the standard for claude-mods shell scripts (docs/TERMINAL-DESIGN.md), with the enclosing panel as the default grammar. The github-ops audit family (repo-scorecard.sh, check-security-posture.sh, check-issues.sh) now sources skills/_lib/term.sh and wraps its human output in the full term_panel_open … term_panel_close frame — brand header, │ body rail, term_section sub-headers, colored term_mark rows, a score pip-bar, and a footer health indicator — matching the fleet-ops look. The --json/data product on stdout stays plain (stream separation preserved; verified zero ANSI on stdout). Every glyph falls back to ASCII under TERM_ASCII=1 (a full scorecard renders pure-ASCII), and color follows the stderr TTY so piping --json | jq keeps framing colored.
• term.sh additions: term_init takes an optional fd (term_init 2) so stream-separated tools detect color on the stream the human actually sees; new term_panel_line (generic rail body-row, the open-ended counterpart to the branch-shaped term_leaf_line), term_mark <ok|bad|warn|skip|na|unknown> checklist primitive, a TERM_ARROW pointer glyph, and github-ops/audit/supply-chain/net-ops brand glyphs — all with registered ASCII proxies. github-ops test suite gained 6 assertions (source-check + ASCII-fallback purity across panel + checklist primitives); 40/40 offline.

Fixed (docs)

• SKILL-SUBAGENT-REFERENCE.md was self-contradictory and misleading (surfaced by external PR #12): it declared "no other top-level keys are permitted" and its validation awk flagged when_to_use/argument-hint/effort as violations — yet those are documented Claude Code top-level fields the repo uses deliberately. Rewrote it to document two layers (the portable Agent Skills six-field minimum vs Claude Code's top-level superset), with a referenced field table, the precedence rule (Claude Code is our target → superset fields stay top-level), and an explicit warning that burying them under metadata disables them. Validation snippet now allowlists the superset and defers to claude plugin validate.

Changed (docs review)

• Refreshed WORKFLOWS.md with a v3.0 orientation banner (skills-first; subagents are isolation/worker-only); fixed naming-conventions.md frontmatter example (metadata block, not top-level arrays); dated RESERVED-COMMANDS.md; noted anthropic-skills-guide.md as a vendored reference pointing at the creation protocol.
• Archived completed-migration docs to docs/archive/: AGENT-SKILLS-COMPLIANCE-BRIEF.md (spec migration done, CI-enforced) and COMMAND-SKILL-PATTERN.md (command→skill conversion done). docs/ top level is now 8 load-bearing docs + archive/ + references/.

Added

• github-ops repo-scorecard.sh (the audit capstone) - one read-only command for a scored repo-health report, fleet-aware. Orchestrates the existing auditors (check-security-posture.sh, check-issues.sh) rather than re-implementing them, and rolls five dimensions — security (w35), metadata (w25), release-consistency (w15), open-issues (w15), latest-Actions-run (w10) — into a 0–100 score + A–F grade with the top-3 fixes per repo. --org sweeps every non-archived repo into a matrix + roll-up (avg/median, worst repos, fleet open-alert total); --min-score N is a CI gate (exit 10 below N). An unreadable dimension scores zero ("n/a"), never a false-healthy. Read-only (CI-asserted); +15 test assertions (34 total). Now the headline of github-ops audit mode.
• github-ops security-posture auditor - scripts/check-security-posture.sh: read-only audit of a repo's GitHub security settings (Dependabot alerts + security updates, secret scanning + push protection, code scanning, private vulnerability reporting, SECURITY.md, default-branch protection). Three things make it more than a toggle-checker: visibility-aware severity (public-repo scanning gaps are findings; private-without-GHAS is a note, not a nag), the exposure layer (where a scanner is enabled it fetches open-alert counts + max severity — the real signal), and an --org fleet sweep that audits every non-archived repo you own in one pass. Emits the exact enable commands but never applies them (a CI-asserted read-only guarantee). Ships assets/SECURITY.md.template; wired into audit mode; +13 offline test assertions (19 total). Exit 10 on gaps/open-alerts, 7 when unavailable.
• github-ops open-issue awareness - scripts/check-issues.sh surfaces open issues you may not have seen (externally-authored + stale), read-only via gh issue list. Wired into the pre-push gate (push-gate/preflight.sh) as a post-gate advisory step: every push to a GitHub remote now flags unseen external/stale issues for that repo. Timeout-bounded, never changes the gate verdict, silent when gh is absent/unauthed or the remote isn't GitHub. Exit 10 = issues to look at, 7 = unavailable (advisory). github-ops gains a 6-assertion offline test suite.
• okf-ops skill - assess, validate, and adopt the Open Knowledge Format (OKF) across markdown+frontmatter knowledge bases. assess-okf.py (read-only) scans a doc tree for OKF-readiness — frontmatter coverage, type presence, a key/value histogram, and a readiness % — so you can find good adoption candidates among many repos; check-okf.py validates a bundle for conformance (hard rules only, honouring OKF's permissive-consumption contract; --strict for CI gating). Honest scope baked in: OKF is a v0.1 draft, adopt per-repo not blanket. Both tools built to the Skill Resource Protocol; OKF format reference + copy-ready concept template; 10-assertion offline self-test.
• adr-ops skill - Architecture Decision Records as a cross-project workflow, generalized from a mature in-house ADR protocol: when-to-write / when-NOT decision rule, the canonical format (BLUF-first ## Decision, fixed section order, frontmatter field set), the proposed→accepted→superseded/deprecated lifecycle, and append-only supersession discipline. Five tools to the Skill Resource Protocol:
  • adr-init.sh - bootstrap ADRs in a repo adopting them cold (dir + lint-clean ADR-001 + generated README)
  • adr-new.sh - scaffold the next sequential ADR (atomic, no-clobber, --apply-supersede flips the superseded record's frontmatter)
  • adr-index.sh - read-only index table from frontmatter; --output writes a generated Markdown index
  • adr-touching.py - query the touches: discovery surface ("what ADRs govern this path before I change it?"); exit 10 when a governing ADR exists, a usable pre-edit/CI guard
  • adr-lint.py - validates required fields, status enum, numbering, section order, cross-file supersession bidirectionality, lifecycle consistency (status vs superseded-by), and stale-touches paths Includes a CI-integration section (gate adr-lint --strict on exit 10). 72-assertion offline self-test.

3.0.0 - 2026-06-10

Added (media stack)

• ytdlp-ops skill - yt-dlp as the media ACQUISITION layer feeding ffmpeg-ops: format selection doctrine (-S sort over -f filters, codec targeting that avoids post-download transcodes), --download-sections clip-at-download, audio-only STT extraction (-x --audio-format opus = stream copy), playlist + --download-archive incremental channel syncs (--break-on-existing --lazy-playlist cron pattern), cookies/auth (--cookies-from-browser, Chrome 127+ Windows caveat, ban avoidance), rate limiting/politeness, SponsorBlock mark-vs-remove, output-template conventions ([%(id)s], byte-safe .100B truncation), subtitles-as-cheap- transcripts, remux-vs-recode doctrine, livestream/premiere capture (--live-from-start, --wait-for-video), batch dry-runs (--print filename), a beyond-YouTube note, and a failure-triage ladder (the nsig/403/429/geo classes incl. TLS-fingerprint blocks → --impersonate, and the EJS class: missing formats from no JS runtime → deno default / --js-runtimes node opt-in, surfaced by the verifier as a warning; "outdated yt-dlp" is the diagnosis for most). Completes the acquire → process chain with ffmpeg-ops. Ships a §7 staleness verifier (check-ytdlp-version.sh: --offline structural in PR CI; --live = installed version >60 days behind the latest GitHub release, a documented core flag vanished from yt-dlp --help, or smoke-extraction failure → exit 10, network unreachable → exit 7 advisory; wired into tests/check-resources.sh + freshness.yml). 6 references, 1 date-stamped preset asset, 28-assertion offline self-test (age logic exercised via test seams - no network in tests).
• ffmpeg-ops skill - probe-first ffmpeg/ffprobe operations: ~30-command cookbook with footgun table (seek/keyframe semantics, yuv420p+faststart, quoting, VFR), EDL-driven editing (edit-as-code: schema asset + cut-from-edl.py, dry-run by default), .cube LUT grading with human-picks-the-grade chooser (gen-luts.py), STT/Whisper prep + the transcript-JSON contract, silence/scene segmentation (detect-segments.py), VMAF/SSIM quality gates (quality-compare.py), two-pass loudnorm automation, hw-encoder proof-encoding (capability-scan.sh - listed ≠ working), chapter authoring from scene/silence detection (make-chapters.py - ffmetadata mux / YouTube description / WebVTT), probe --doctor triage (each hazard - VFR, HDR transfer, rotation, interlacing, non-yuv420p, moov-at-EOF - paired with its exact fix command, exit 10), target-size compression (smart-compress.py - computed two-pass bitrate, auto audio/downscale, size-verified), scrub-preview sprites + WebVTT thumbnail track (make-sprites.py), an error-decoder reference (cryptic message → cause → fix), and a §7 staleness verifier (verify-commands.sh, wired into PR CI + freshness). Color grading is a first-class wing: a ~40-recipe look catalog (film stocks incl. CineStill halation as a verified composite, signature movie grades, era/genre moods, Sin City colorhold) with per-look scope checks and failure modes, an 18-variant mono/duo/tritone tone-map family (chroma = stop distance from the grey axis), the Hald-CLUT grade-anywhere→LUT workflow, a scope-matching ladder with its governing rule (transfer the chroma fingerprint globally; match key per scene-type, never the global mean) and a real-footage worked extraction (grimdark), plus a skin-tone equity caveat verified on the Kodak test portraits. gen-luts.py carries 32 parametric looks (channel-mix + 2/3-stop gradient maps). 19 references, 3 assets, 107-assertion self-test with lavfi-synthesized fixtures (no binary fixtures in repo).

Fixed (media stack)

• ffmpeg-ops/cut-from-edl.py (found by real-media E2E): the output directory was created after ffmpeg opened the temp output, so any -o into a not-yet-existing directory died with a cryptic "Error opening output files"; and CLI -o resolved against the EDL's directory instead of the CWD (-o work/final.mp4 with the EDL in work/ silently meant work/work/final.mp4). -o is now CWD-relative (the EDL's own output field stays EDL-relative per the schema), and the destination dir is created before the concat runs.

Added (skill resource protocol)

• docs/SKILL-RESOURCE-PROTOCOL.md - the build standard for skill scripts/, assets/, and references/: stream separation, semantic exit codes, --help with EXAMPLES, first-comment-block contract, --json envelopes, agent safety, the resource-scaffold checklist, and the staleness-verifier pattern (an --offline structural check that gates PR CI plus a --live drift check that runs scheduled, never blocking a PR on a network blip)
• Four verifier/scanner scripts built to the protocol: claude-api-ops/check-model-table.py (model+pricing table drift), terraform-ops/check-action-refs.sh (GitHub Action uses: refs resolve — catches the exact trivy-action tag bug from v3.0), claude-code-ops/validate-hooks-json.py (lint a hooks.json against the 30-event catalog), playwright-ops/triage-flakes.py (rank flaky tests from a JSON report). Plus assets: agentic-loop.py, output-schema.json, hooks.json.template
• CI: tests/check-resources.sh runs the offline verifiers in PR CI; .github/workflows/freshness.yml runs the live drift checks weekly (advisory)

Removed

• 20 expert agents deprecated as part of the skills-first restructure (23 → 3 agents):
  • 11 language/framework experts → their -ops skill twins (python, typescript, javascript, go, rust, react, vue, astro, laravel, sql, postgres)
  • cypress-expert → cypress-ops; cloudflare-expert + wrangler-expert → cloudflare-ops; bash-expert → bash-ops; craftcms-expert → craftcms-ops; payloadcms-expert → payloadcms-ops; asus-router-expert → asus-router-ops
  • claude-architect → folded into claude-code-ops; aws-fargate-ecs-expert → folded into container-orchestration Per Anthropic's guidance, knowledge belongs in skills (progressive disclosure, single source of truth); subagents are reserved for context isolation. The only agents that remain are pure isolation/worker roles: git-agent, firecrawl-expert, project-organizer. Dispatching skills route general-purpose agents with skill preloading.
• claude-code-debug, claude-code-headless, claude-code-hooks skills - merged into claude-code-ops (content was written against Claude Code ~2.0; the stale $TOOL_INPUT hook contract is gone, stdin JSON is current)

Added

• claude-api-ops skill - building ON Claude: Messages API, tool use, prompt caching, structured outputs (output_config.format), batches, extended thinking, model selection, Agent SDK (Python + TypeScript)
• playwright-ops skill - e2e testing: selector hierarchy, fixtures/POM, network mocking, auth storageState, CI sharding, flake hunting, config template
• terraform-ops skill - Terraform/OpenTofu IaC: state management, module patterns, OIDC CI/CD workflow template, drift detection, write-only secrets, native terraform test
• claude-code-ops skill - merges + refreshes claude-code-debug, claude-code-headless, claude-code-hooks against current docs: 30-event hook catalog with JSON contracts, current skill frontmatter spec, headless/CLI reference, extension debugging decision trees (+ extension-architecture from claude-architect)
• cypress-ops, cloudflare-ops, bash-ops skills - converted from the cypress/cloudflare/wrangler/bash agents and refreshed against current docs (Cypress data-test/Test Replay/cy.session; wrangler deploy not publish, jsonc config, Workers static assets; defensive bash to the resource protocol)
• craftcms-ops, payloadcms-ops, asus-router-ops skills - converted from the niche CMS/router agents and refreshed against current docs (Craft 5 Matrix-as-entries; Payload 3 Next.js-native + Local API; Asuswrt-Merlin hardening + WireGuard)
• Live security guard hooks: config-change-guard.sh (ConfigChange event - scans edited Claude settings files for worm-persistence IOCs the moment they're written, reusing integrity-audit patterns) and worktree-guard.sh (PreToolUse - mechanically enforces rules/worktree-boundaries.md)
• Plugin hook auto-wiring (hooks/hooks.json) - plugin installs get the security-advisory hook set (pre-install-scan, manifest-dep-scan, session-start unicode scan, config-change guard, worktree guard) with zero hand-wiring; formatting/lint hooks stay opt-in examples
• fleet track command - register natively-spawned branches as fleet lanes
• New frontmatter on high-traffic skills: when_to_use (10 skills), argument-hint (iterate/review/testgen/explain), effort: high (iterate/review)
• README "Skill Description Budget" guidance - /doctor overflow check, skillOverrides, 1,536-char per-skill cap
• CI: doc-drift gate (tests/doc-drift.sh) - docs must match disk
• CI: skill behavioural test suites (tests/run-skill-tests.sh)

Fixed

• fleet.sh ensure_fleet_dir returned 1 under set -e on every invocation after the first, silently killing post-init commands
• fleet-ops e2e suite asserted a worktree path fleet.sh no longer uses (now 29/29 against real behaviour)

Changed

• fleet-ops v2 - repositioned as landing discipline (queue, test gate, pre-land scrub, one-shot revert) on top of native agent teams / background agents, which now own the spawning half; no longer EXPERIMENTAL except the daemon
• /save + /sync repositioned - native auto-memory covers single-machine context; these commands' value is portable state: task restore, git-trackable, team-shareable, cross-machine
• supply-chain-defense description trimmed under the 1,536-char listing cap
• README/AGENTS.md/PLAN.md reconciled with actual inventory; ghost references removed (rules/thinking.md, docs/DASH.md)
• tests/skills/functional/git-workflow.* renamed to git-cli-tools.*

2.10.1 - 2026-05-29

Fixed

• Plugin + marketplace manifests made valid against the official schema; claude plugin validate added as a CI gate (#4)

2.10.0 - 2026-05-25

Added

• prompt-injection-defense skill - instruction-integrity defense: hidden-Unicode scanner (bidi/Trojan Source, tag-block smuggling, zero-width), byte-faithful sanitizer, SessionStart + git pre-commit hooks, rules/prompt-injection.md

2.9.0 - 2026-05-25

Added

• supply-chain-defense skill - behavioural-first dependency security: Socket.dev integration (free CLI + zero-auth depscore MCP), exposure-check across 6 ecosystems + editor extensions, integrity-audit for worm persistence, 7-day release cooldown, install + manifest advisory hooks, rules/supply-chain.md, 42-assertion offline test suite

2.8.0 - 2026-05-18

Added

• mac-ops skill finalized - macOS workstation diagnostics, peer to windows-ops: 23 scripts + 11 references (TCC privacy, wake reasons, Spotlight, APFS storage pressure)

[2.7.0] - [2.7.8] - 2026-05-17 to 2026-05-18

Added

• mac-ops incremental build-up: kext/firewall/keychain/bluetooth/font audits, brew-health, sysdiagnose-helper, quickrun consolidator, worked examples

2.6.0 - 2026-05-15

Added

• windows-ops skill - Windows workstation diagnostics: health-audit panel, crash-triage (Event 41 BugCheck decoding), recover-clone for dying drives

2.5.0 - 2026-05-14

Added

• net-ops skill - cross-platform network troubleshooting ladder (link → app), IPv6 classifier, MTU/PMTU, DoH detection, --watch/--json/--redact
• portless-ops skill - local-dev HTTPS proxy operations for Vercel Labs portless
• process-compose-ops skill - Process Compose supervisor operations

Fixed

• summon + fleet-ops registered in plugin manifest (were committed but unindexed)

Removed

• /canvas command + canvas-tui package - experimental, Warp-specific, unused; removes the only npm runtime-dep surface

[2.4.12] - 2026-05-05

Fixed

• install.sh made cross-platform (Linux/macOS/Windows Git Bash)

[2.4.11] - 2026-05-02

Added

• summon skill - transfer Claude Desktop Code-tab sessions between accounts

[2.4.10] - 2026-04-29

Changed

• github-ops Recent Updates rule sharpened: features-not-bugs criteria

[2.4.9] - 2026-04-26

Added

• git-ops hygiene checks - status.sh flags feature-branch checkouts, stale merges
• docs/references/claude-desktop-internals.md - Desktop session architecture map

[2.4.7] - 2026-04-26

Fixed

• push-gate first-push to new remote (gitleaks scan branches on remote-ref existence)

[2.4.6] - 2026-04-26

Added

• github-ops skill - GitHub remote operations: repo creation, releases, metadata, README Recent Updates convention; three modes (new/update/audit)

[2.4.5] - 2026-04-26

Added

• leveldb-ops skill - read Chromium/Electron LevelDB stores via ccl_chromium_reader

[2.4.4] - 2026-04-25

Changed

• /iterate enhancements - Batch+bisect, Until/Stagnation stop conditions, branch isolation, iterate/best tag, always-summarize-on-exit

[2.4.3] - 2026-04-24

Added

• Worktree-aware git-ops (status.sh + worktree-survey.sh)
• push-gate skill - pre-push secret/forbidden-file gate, no bypass
• rules/worktree-boundaries.md

Fixed

• auto-skill suggestions persisted to pending.log, surfaced at /sync

[2.4.2] - 2026-04

Changed

• Registered push-gate, auto-skill visibility fix

[2.4.1] - 2026-04

Added

• 8 daemon output styles (Atlas, Coach, Harbour, Meridian, Noir, Roast, Sage, Scout) - 13 total

[2.4.0] - 2026-04

Added

• auto-skill skill - self-learning skill creation via PostToolUse/Stop hooks
• pigeon skill (renamed from agentmail) - inter-session pmail, SQLite-backed

[2.3.1] - 2026-04

Added

• genart-ops skill (1,843 lines)

Changed

• All skills migrated to the Agent Skills specification (agentskills.io)

[2.3.0] - 2026-03

Added

• Orchestrator-dispatch pattern: git-ops + git-agent (replaces git-workflow), perf-ops, security-ops parallel audits
• Skill preloading for dispatched agents; model: sonnet for expert dispatch

[2.2.x] - 2026-03

Changed

• /introspect Session Insights; /setperms 74 default permissions

Removed

• claude-code-templates (redundant with first-party skill-creator)

[2.1.0] - 2026-03

Added

• /iterate skill - autonomous improvement loop (Karpathy autoresearch pattern)

[2.0.0] - 2026-03

Added

• 22 new -ops skills (React, Vue, Go, Rust, TypeScript, Docker, CI/CD, PostgreSQL, Nginx, Auth, Monitoring, Debug, MCP, Tailwind, and more)
• cc-session CLI, 3 hooks, 5 output styles

Changed

• All 14 -patterns skills renamed to -ops

[1.x] - 2025-11 to 2026-02

Added

• Initial toolkit: session continuity (/save + /sync, schema v3.1), expert agents, Python skill family, tech-debt scanner, modern CLI toolkit, validation suite