2026-03-26 - 2026-09-26
Overview
3 Releases published by 1 user
Published
v2.0.1
v2.0.1 — context-set trim and published metadata
Published
v2.0.0
v2.0.0 — ADR-0020 context contract
Published
v1.0.0
v1.0.0
35 Pull requests merged by 2 users
Merged
#142 chore: migrate git host from git.dev.rkdr.net to git.rkdr.net
Merged
#141 fix(tests): exclude build/ from bats test discovery
Merged
#140 fix(gates): check body-level routing targets, not just descriptions
Merged
#139 feat(kyberforge): make Research doc name one Research registry
Merged
#138 fix(research): restore subagent fan-out (#116)
Merged
#137 fix(gates): waive lockfile-exists for a package, which is not an install root
Merged
#136 feat(onedev): redistribute TOD's agent skills through a plugin wrapper
Merged
#135 refactor!: carry out the simplification audit across gates, tests, plugins and docs
Merged
#134 docs: trim four duplicated session rules from AGENTS.md
Merged
#133 refactor(bin): move non-spec root files into references/ across four skills
Merged
#131 fix(tests): pin test-check-rtk-prefix.sh's pre-#113 corpus to a fixed SHA
Merged
#130 Close four small ADR-0020-retrofit follow-ups (#125, #127, #113, #118)
Merged
#129 refactor(skills): retrofit the corpus to the ADR-0020 context contract
Merged
#106 fix(gitea): correct which pull_request_write params apply on create
Merged
#105 docs: cut the every-session context set by 74%, and fix what the cut broke
Merged
#103 feat(kyberforge): ADR-0020 context contract for skills and agents
Merged
#98 feat(apm): consume plugins through apm and keep the install fresh at SessionStart
Merged
#95 feat(kyberforge): execute the plugin→APM conversion
Merged
#93 feat(kyberforge): retarget forge skills to author/audit APM content
Merged
#91 feat(kyberforge): add apm-install/apm-workflow/apm-orchestrate, ADR-0015
Merged
#85 feat(lint): wire Vale as deterministic prefilter for skill-audit/agent-audit
Merged
#86 docs(agents-md): shrink AGENTS.md and prefer plugin skills over shell
Merged
#83 chore(settings): enable gitea plugin
Merged
#82 chore(gitea): remove old flat gitea skill, superseded by plugins/gitea
Merged
#81 docs(agents): add setup, testing, and commit conventions
Merged
#80 feat(core): add AGENTS.md authoring/review tooling
Merged
#67 refactor(gitea): deep modules — split flat dispatch skill into 6 domain skills + orchestrator + agent
Merged
#77 fix(tests): auto-init submodules when bats binary is missing
Merged
#76 docs(agents): document worktree/branch cleanup as part of PR close-out
Merged
#74 fix(kyberforge): require git-log commit verification and forbid self-spawned rechecks
Merged
#73 docs(agents): document fork task-scope and TaskList access constraints
Merged
#72 fix(install): resolve git hooks dir via git plumbing, fix stale plugin manifests
Merged
#65 fix(factory): embed lessons from #63 into gitea and kyberforge skills
Merged
#64 fix(agents): relocate provenance sources.md outside agents/ dir
Merged
#7 chore: move gitea to bin
92 Issues closed from 1 user
Closed
#124 fix(gates): body-level routing targets are checked by nothing
Closed
#53 feat: CI gate for /improve-codebase-architecture
Closed
#92 gitea-mcp: pull_request_review_write reply_comment returns HTTP 405 on this instance
Closed
#66 gitea: wire real MCP server config into plugins/gitea/.mcp.json
Closed
#121 provenance: Research doc: slug and path conventions do not match what validate-provenance expects
Closed
#116 design(bin): research had a fan-out design its allowed-tools never permitted
Closed
#101 refactor(kyberforge): merge skill-audit + agent-audit, and address the author-pair duplication
Closed
#123 design: nothing in the 39-skill corpus produces a PRD
Closed
#122 refactor(bin): four skills still keep non-spec files at the skill root
Closed
#55 feat: tests/test-pre-commit-installed.sh
Closed
#113 git plugin: rtk git vs bare git is inconsistent across skills
Closed
#118 validate-provenance.sh passes a source_keys entry that asserts a contribution which never happened
Closed
#125 fix(kyberforge): description-quality.md lists user-invocable: false as a hand-invocation marker, which is backwards
Closed
#127 decide: per-skill metadata.version has no owning rule
Closed
#112 git plugin: cherry-pick and revert have three claimants
Closed
#117 vale: the Kyberforge style is scoped to SKILL.md, so every references/ file is unlinted
Closed
#119 research drift: installation.md omits apm runtime status and apm runtime remove, which exist in apm 0.28.0
Closed
#126 feat(agent-audit): implement ADR-0016's reserved SUGGESTION for unenforceable tool-restriction claims
Closed
#128 design(core): no provider-file audit path — "is my CLAUDE.md OK?" reaches a write-only skill
Closed
#120 fix(apm-orchestrate): description/delegation FAILs and confirmed add-package mapping bug
Closed
#114 refactor(bin): prototype's two flow files sit at the skill root instead of references/
Closed
#111 validate-provenance.sh: no way to declare a reference file as house-authored
Closed
#110 skill-size-check: a boundary clause naming a dotted filename is invisible to the gate
Closed
#108 fix(gates): boundary-clause check fires on hand-invoked skills, which are contractually exempt
Closed
#107 fix(gates): boundary-target resolver checks only the first target per arrow clause
Closed
#99 refactor(skills): retrofit the 39 skills to the ADR-0020 context contract
Closed
#115 fix(core): validate-adapter.sh's --no-import-syntax flag is a no-op
Closed
#109 docs(kyberforge): body-discipline.md and the ADR's cited dispatch exemplar cannot both be satisfied
Closed
#100 fix(skills): four description boundary clauses route to targets that do not exist
Closed
#104 fix(gitea): gitea-prs reference wrongly claims milestone is settable on PR create
Closed
#78 post-push hook never ran: git has no such hook, so no plugin cache was refreshed
Closed
#90 Execute the plugin→APM conversion
Closed
#97 Residual low-severity items from PR #95's fourth review round
Closed
#96 Guard unguarded array expansion in statusline-command.sh (latent, ships to users)
Closed
#89 Adapt skill-author/skill-audit/agent-author/agent-audit to author/audit APM-native (.apm/) content
Closed
#88 Build agents/skills to execute marketplace→APM conversion
Closed
#47 0028 — Chunk 3 closure: update skills-index, update spec, behavioral tests
Closed
#50 fix: gitleaks.toml — allowlist false positive blocking commits (U1)
Closed
#54 feat: plugin validate in pre-push hook + tests/test-plugin-validate.sh
Closed
#56 feat: tests/test-inventory-crossrefs.sh — zero ../ cross-reference warnings
Closed
#87 Cut first release tag (vX.Y.Z) once PR #85 merges to main
Closed
#84 Add Vale as deterministic pre-filter for skill-audit/agent-audit
Closed
#79 Add AGENTS.md authoring/review tooling (agentsmd-author, agentsmd-audit, provider-adapter-author)
Closed
#6 refactor(kyberforge/gitea): deep modules — move how into reference files, SKILL.md becomes orchestrator
Closed
#75 agents: coordinator-created worktrees and branches must be cleaned up automatically when their PR merges
Closed
#71 kyberforge:skill-author must not self-spawn its own independent-recheck subagent
Closed
#69 kyberforge skill/agent authoring should require git log verification, not just git diff, before reporting a task done
Closed
#70 document that fresh (non-fork) subagents cannot access TaskGet/TaskUpdate tools
Closed
#68 agents: forks must not autonomously continue past their assigned task when a shared task list exists
Closed
#63 agents/sources.md convention conflicts with claude plugin validate --strict
Closed
#62 feat(git-plugin): Build complete git workflow automation suite
Closed
#15 refactor: align repo with marketplace/plugin model — migrate docs and issues to Gitea, rewrite specs, clean up stale artifacts
Closed
#61 Add forge skill to kyberforge — orchestrates grill + artifact-type routing
Closed
#48 feat: role agents — Architect, Developer, Reviewer, Security, QA, Ops (core/agents/)
Closed
#52 research: QA agent design — what does automated agent testing look like in practice?
Closed
#57 feat: tests/run-all-tests.sh — single test entry point
Closed
#58 feat: extend test-governance-layer.sh with structural enforcement checks
Closed
#59 feat: bidirectional reference scanner script
Closed
#60 kyberforge: agent-audit provenance/sources chain validation
Closed
#11 kyberforge: agent-audit skill (agent-author follow-ons)
Closed
#28 0017 — factory/write-eval (bootstrap skill)
Closed
#31 0018 — factory/write-skill (bootstrap skill)
Closed
#24 0016 — Second grill: skill implementation workflow
Closed
#22 0015 — AGENTS.md refactor (prerequisite)
Closed
#42 0014 — docs/spec/ and VISION.md refactor
Closed
#40 0013 — LESSONS.md for this repo
Closed
#38 0012 — Governance manual test plan
Closed
#35 0011 — Governance reference doc updates
Closed
#33 0010 — Governance supporting docs
Closed
#32 0009 — governance.md and @import wiring
Closed
#30 0008 — Restructure docs/ subdirectories and migrate existing PRD
Closed
#29 0007 — Write core/instructions/testing.md
Closed
#27 0006 — Write core/instructions/git.md
Closed
#26 0005 — Write core/instructions/coding.md
Closed
#25 0004 — Rewrite providers/claude-code/CLAUDE.md and retire global.md
Closed
#23 0003 — Claude Code status line
Closed
#21 0002 — install.sh — deploy script
Closed
#20 0001 — Repo skeleton: content files
Closed
#19 PRD: Chunk 3 — Skills Library Rebuild
Closed
#18 PRD: Governance Instruction Layer (Phase 1)
Closed
#17 PRD: Chunk 2 — Core Instructions
Closed
#16 PRD: Chunk 1 — Repo Skeleton + install.sh
Closed
#14 feat: add mattpocock-skills to marketplace and fix version bump gaps across authoring skills
Closed
#13 feat(kyberforge): add plugin-author and marketplace-author skills
Closed
#12 feat(kyberforge): add pc-author and pc-run skills for pre-commit management
Closed
#10 feat(kyberforge): implement agent-author skill
Closed
#8 feat: skill-audit — validate sources provenance chain
Closed
#4 bug: skill-author loses source provenance — research output not recorded in skill references
Closed
#1 Research and configure Gitea provider: workflows, skills, labels, and content rules
Closed
#5 feat(kyberforge): merge skill-write and skill-improve into skill-author
Closed
#3 forge agent — session handoff 2026-06-21 (redo in fresh session)
Closed
#2 audit: full test-run — 3 fixes applied, 4 untracked items, kyberforge improvements needed
106 Issues created by 1 user
Opened
#1 Research and configure Gitea provider: workflows, skills, labels, and content rules
Opened
#2 audit: full test-run — 3 fixes applied, 4 untracked items, kyberforge improvements needed
Opened
#3 forge agent — session handoff 2026-06-21 (redo in fresh session)
Opened
#4 bug: skill-author loses source provenance — research output not recorded in skill references
Opened
#5 feat(kyberforge): merge skill-write and skill-improve into skill-author
Opened
#6 refactor(kyberforge/gitea): deep modules — move how into reference files, SKILL.md becomes orchestrator
Opened
#8 feat: skill-audit — validate sources provenance chain
Opened
#9 feat: skill refresh workflow — re-research, diff, propagate, re-provenance
Opened
#10 feat(kyberforge): implement agent-author skill
Opened
#11 kyberforge: agent-audit skill (agent-author follow-ons)
Opened
#12 feat(kyberforge): add pc-author and pc-run skills for pre-commit management
Opened
#13 feat(kyberforge): add plugin-author and marketplace-author skills
Opened
#14 feat: add mattpocock-skills to marketplace and fix version bump gaps across authoring skills
Opened
#15 refactor: align repo with marketplace/plugin model — migrate docs and issues to Gitea, rewrite specs, clean up stale artifacts
Opened
#16 PRD: Chunk 1 — Repo Skeleton + install.sh
Opened
#17 PRD: Chunk 2 — Core Instructions
Opened
#18 PRD: Governance Instruction Layer (Phase 1)
Opened
#19 PRD: Chunk 3 — Skills Library Rebuild
Opened
#20 0001 — Repo skeleton: content files
Opened
#21 0002 — install.sh — deploy script
Opened
#22 0015 — AGENTS.md refactor (prerequisite)
Opened
#23 0003 — Claude Code status line
Opened
#24 0016 — Second grill: skill implementation workflow
Opened
#25 0004 — Rewrite providers/claude-code/CLAUDE.md and retire global.md
Opened
#26 0005 — Write core/instructions/coding.md
Opened
#27 0006 — Write core/instructions/git.md
Opened
#28 0017 — factory/write-eval (bootstrap skill)
Opened
#29 0007 — Write core/instructions/testing.md
Opened
#30 0008 — Restructure docs/ subdirectories and migrate existing PRD
Opened
#31 0018 — factory/write-skill (bootstrap skill)
Opened
#32 0009 — governance.md and @import wiring
Opened
#33 0010 — Governance supporting docs
Opened
#34 0019 — Factory skills: write-adr, write-issue-spec, write-workflow, upgrade-skill, validate-skill
Opened
#35 0011 — Governance reference doc updates
Opened
#36 0020 — Design skills: grill-lean, grill-me, write-prd, architecture-review, break-into-issues, prototype
Opened
#37 0021 — Implement skills: implement-feature, tdd, refactor, diagnose
Opened
#38 0012 — Governance manual test plan
Opened
#39 0022 — Test skills: write-tests, generate-test-data, review-test-coverage
Opened
#40 0013 — LESSONS.md for this repo
Opened
#41 0023 — Review skills + cliff.toml: code-review, security-review, pr-description, changelog-entry
Opened
#42 0014 — docs/spec/ and VISION.md refactor
Opened
#43 0024 — Deploy skills: write-ci-pipeline, write-deployment-config, write-ai-review-workflow, deployment-checklist
Opened
#44 0025 — Operate skills: write-runbook, incident-diagnosis, post-mortem, inspect-deployment
Opened
#45 0026 — IaC skills: write-docker-compose, iac-security-review
Opened
#46 0027 — Cross-cutting skills: session-handoff, governance-check, git-commit-message, improve-codebase-architecture, triage, zoom-out, caveman
Opened
#47 0028 — Chunk 3 closure: update skills-index, update spec, behavioral tests
Opened
#48 feat: role agents — Architect, Developer, Reviewer, Security, QA, Ops (core/agents/)
Opened
#49 research: agent confirmation model — define autonomy-first threshold
Opened
#50 fix: gitleaks.toml — allowlist false positive blocking commits (U1)
Opened
#51 feat: CI enforcement pipeline — secret scanning, licence scanning, dep scanning, approval gate, audit logging
Opened
#52 research: QA agent design — what does automated agent testing look like in practice?
Opened
#53 feat: CI gate for /improve-codebase-architecture
Opened
#54 feat: plugin validate in pre-push hook + tests/test-plugin-validate.sh
Opened
#55 feat: tests/test-pre-commit-installed.sh
Opened
#56 feat: tests/test-inventory-crossrefs.sh — zero ../ cross-reference warnings
Opened
#57 feat: tests/run-all-tests.sh — single test entry point
Opened
#58 feat: extend test-governance-layer.sh with structural enforcement checks
Opened
#59 feat: bidirectional reference scanner script
Opened
#60 kyberforge: agent-audit provenance/sources chain validation
Opened
#61 Add forge skill to kyberforge — orchestrates grill + artifact-type routing
Opened
#62 feat(git-plugin): Build complete git workflow automation suite
Opened
#63 agents/sources.md convention conflicts with claude plugin validate --strict
Opened
#66 gitea: wire real MCP server config into plugins/gitea/.mcp.json
Opened
#68 agents: forks must not autonomously continue past their assigned task when a shared task list exists
Opened
#69 kyberforge skill/agent authoring should require git log verification, not just git diff, before reporting a task done
Opened
#70 document that fresh (non-fork) subagents cannot access TaskGet/TaskUpdate tools
Opened
#71 kyberforge:skill-author must not self-spawn its own independent-recheck subagent
Opened
#75 agents: coordinator-created worktrees and branches must be cleaned up automatically when their PR merges
Opened
#78 post-push hook never ran: git has no such hook, so no plugin cache was refreshed
Opened
#79 Add AGENTS.md authoring/review tooling (agentsmd-author, agentsmd-audit, provider-adapter-author)
Opened
#84 Add Vale as deterministic pre-filter for skill-audit/agent-audit
Opened
#87 Cut first release tag (vX.Y.Z) once PR #85 merges to main
Opened
#88 Build agents/skills to execute marketplace→APM conversion
Opened
#89 Adapt skill-author/skill-audit/agent-author/agent-audit to author/audit APM-native (.apm/) content
Opened
#90 Execute the plugin→APM conversion
Opened
#92 gitea-mcp: pull_request_review_write reply_comment returns HTTP 405 on this instance
Opened
#94 kyberforge: add authoring/audit support for hooks, instructions, and prompt primitives
Opened
#96 Guard unguarded array expansion in statusline-command.sh (latent, ships to users)
Opened
#97 Residual low-severity items from PR #95's fourth review round
Opened
#99 refactor(skills): retrofit the 39 skills to the ADR-0020 context contract
Opened
#100 fix(skills): four description boundary clauses route to targets that do not exist
Opened
#101 refactor(kyberforge): merge skill-audit + agent-audit, and address the author-pair duplication
Opened
#102 build: no gate asserts that tool-owned paths stay out of autofixing formatter scope
Opened
#104 fix(gitea): gitea-prs reference wrongly claims milestone is settable on PR create
Opened
#107 fix(gates): boundary-target resolver checks only the first target per arrow clause
Opened
#108 fix(gates): boundary-clause check fires on hand-invoked skills, which are contractually exempt
Opened
#109 docs(kyberforge): body-discipline.md and the ADR's cited dispatch exemplar cannot both be satisfied
Opened
#110 skill-size-check: a boundary clause naming a dotted filename is invisible to the gate
Opened
#111 validate-provenance.sh: no way to declare a reference file as house-authored
Opened
#112 git plugin: cherry-pick and revert have three claimants
Opened
#113 git plugin: rtk git vs bare git is inconsistent across skills
Opened
#114 refactor(bin): prototype's two flow files sit at the skill root instead of references/
Opened
#115 fix(core): validate-adapter.sh's --no-import-syntax flag is a no-op
Opened
#116 design(bin): research had a fan-out design its allowed-tools never permitted
Opened
#117 vale: the Kyberforge style is scoped to SKILL.md, so every references/ file is unlinted
Opened
#118 validate-provenance.sh passes a source_keys entry that asserts a contribution which never happened
Opened
#119 research drift: installation.md omits apm runtime status and apm runtime remove, which exist in apm 0.28.0
Opened
#120 fix(apm-orchestrate): description/delegation FAILs and confirmed add-package mapping bug
Opened
#121 provenance: Research doc: slug and path conventions do not match what validate-provenance expects
Opened
#122 refactor(bin): four skills still keep non-spec files at the skill root
Opened
#123 design: nothing in the 39-skill corpus produces a PRD
Opened
#124 fix(gates): body-level routing targets are checked by nothing
Opened
#125 fix(kyberforge): description-quality.md lists user-invocable: false as a hand-invocation marker, which is backwards
Opened
#126 feat(agent-audit): implement ADR-0016's reserved SUGGESTION for unenforceable tool-restriction claims
Opened
#127 decide: per-skill metadata.version has no owning rule
Opened
#128 design(core): no provider-file audit path — "is my CLAUDE.md OK?" reaches a write-only skill