When the invoking session is the planner, skip the launch-claim and
monitor ceremony and run the loop in place. Reconcile stale pane IDs
instead of treating them as mismatches. Require clean/pushed task
worktrees before merge, review with one independent subagent, and assert
cross-ticket checks on the integration branch. Pair start/finish events
so silent phase gaps are detectable.
The orchestrator no longer monitors a running planner; it confirms startup
and exits. The planner is the only full CLI agent and delegates implementer,
reviewer, fixer, and merger roles to native Pi subagents instead of Herdr- or
CLI-launched agents. Update dispatcher, run-loop, state, and agent interface
copy to match.
Rewrote the skill so the invoking session only discovers an existing
run or launches a fresh planner, then monitors via Herdr read-only
commands. Workers keep state in .orchestrate/<slug>/ with run.json and
events.jsonl; the tracker holds all handoffs and decisions. Re-invoking
the same command recovers monitoring after an orchestrator outage, and
an in-progress label plus atomic launch claims prevent duplicate
planners.
Rename thermo-nuclear-code-quality-review to code-quality-review and drop
the dramatic wording; allow security-audit model invocation. Both now run
as extra axes in the implementation-orchestrator and orchestrate-herdr
review steps.
Rewrite SKILL.md, dispatcher and run-loop references: the skill now takes
<spec-ticket> [worker], always starts fresh agents, and follows the
implement-spec flow (integration branch, frontier tickets, per-task review
and merge) instead of the old plan-and-publish-to-merged-PR pipeline.
Update the engineering README description to match.
Declare display_name, short_description, and disable implicit invocation
for each skill via agents/openai.yaml. Also fix stale /setup-matt-pocock-skills
reference in implement-spec to /setup-skills.
New model-invoked skill + feedback_bundle.py: creates a Gitea issue whose
body links the current HEAD commit and uploads logs/screenshots as issue
assets. Handles label-name-to-ID resolution, tea-config token lookup, and
the disabled-attachments (403/404) case. Index updated in root and
engineering READMEs.
Replace the stale common/ structure docs with the current skills/ bucket
layout. Root and bucket READMEs now index every SKILL.md by invocation
type, AGENTS.md lists the ten actual buckets, and the project context
pack is condensed to purpose, structure, conventions, and validation.
Adds a user-invoked pr skill with a Summary/Evidence/Merge Danger
template, visual-format guidance adapted from Dex Horthy's show-me
skill (attribution in CREDITS.md), and an openai agent interface.
Add an explicitly invoked Pi orchestration skill that drives a plan from the
current conversation or an @file through a published spec, linked dependency-aware
tickets, and one-at-a-time isolated implementation to a merged PR, using Herdr
and Pi agents with no Bun CLI or orchestration runtime.
- orchestrate-herdr (SKILL.md + dispatcher/run-loop/state references): worker-node
selection and Herdr discovery/validation, plan-and-publish, one-at-a-time task
loop (claim, isolate, implement, verify, four-axis review, scoped fix rounds max
three, transient retries max two, reconcile uncertain side effects, escalate),
PR/merge/close gates, authoritative worker-node state, cross-system recovery,
and owned-only cleanup that preserves dirty worktrees and retained state.
- dual-review: self-contained correctness/security + maintainability review axis
(four-axis review with code-review).
- to-spec and to-tickets: drop disable-model-invocation so the orchestration
workflow can invoke them.
- READMEs: list the two new skills.
tea --description does not read stdin via @-, so a piped heredoc silently
creates an issue whose body is the literal "@-". Document the command-
substitution form, a post-create body check, and inspecting tea --help
before relying on CLI syntax.
The name dev-workflow described a category, not what the skill does. It
orchestrates implementation of already-planned tickets, so the new name
says that. Skill dir, docs page, name field and openai display name are
updated to match; validate-skill.sh moves with the dir and its paths are
fixed for the new depth and engineering/ layout.
Add two engineering skills for implementing tracker tickets end to end:
- dev-workflow: selects workflow:ready-for-agent tickets, runs them in
isolated worktrees, opens one PR/MR per ticket, reviews, merges, and
cleans up.
- forge-cli: detects the forge from origin and wraps gh, glab, and tea
for Gitea/Forgejo, GitHub, and GitLab.
Docs cover install, prerequisites, ticket labels, and per-forge CLI usage.
Reorganize README.md with category sections (engineering, pkm, etc.)
for both user-invoked and model-invoked skills. Add new subcategory
READMEs for skill-authoring and thinking-and-docs. Replace
/setup-matt-pocock-skills references with setup-skills across all SKILLs.
Add OpenAI interface metadata to 14 skills (commit-staged,
implement-isolation, implement-isolation-tmux, project-context-pack,
setup-skills, agent-handoff, knowledge-gardener, tmux-launch-agent,
conversation-summary, crit, pkm-curation, research-vault,
youtube-video-capture, lsp-code-analysis).
Add new lsp-code-analysis skill for semantic code navigation and
refactoring via the Language Server Protocol.
Replace the monolithic implement-issue and implement skills with two
focused alternatives: implement-isolation (subagent + reviewer) and
implement-isolation-tmux (tmux-launch-agent for fire-and-forget).
Update the README to list the new skills.
Clean up all skill bucket READMEs — sort entries alphabetically within
user/model sections, add category descriptions to each bucket, simplify
the deprecated forge-skills doc to a bare notice, update triage label
count (5→7) in AGENTS.md, and add a Project Context Pack section.
Captures subtitles from a YouTube video via yt-dlp, produces a prose
summary with key takeaways, and saves both as OKF-conformant notes
into the Video/ bundle in the Obsidian vault.
Add deprecated forge-gitea and forge-github skill entries to the main
README, and create a new misc category for rarely used skills like
tmux-launch-agent.
- Add wayfinding operations section to all issue tracker backends (GitHub, GitLab, Gitea)
- Update skill names from `to-issues`/`to-prd` to `to-tickets`/`to-spec`
- Fix Gitea CLI reference from `gitea` to `tea`
- Fix ADR path from `adr` to `docs/adr`
- Fix various typos and whitespace issues in Gitea tracker docs
- Restructure skill listings across all README files
- Add new skills: agent-handoff, commit, tmux-launch-agent, lsp-code-analysis
- Move knowledge-gardener and agent-handoff to in-progress status
- Reassign skills between user-invoked and model-invoked categories
- Update descriptions to include OKF v0.1 conventions for conversation-summary and research-vault
- Remove empty productivity skill bucket
- Consolidate deprecated skill sections
Remove `docs/agents/agent-cli.md` and all references to it from
`setup-skills/SKILL.md` (contents list, docs files list, and the
Agent CLI block template). Fix typo "languagee" → "language" in
`setup-skills/SKILL.md` and "labels" → "label" in
`implement-issue/SKILL.md`.
- New common/misc/tmux-launch-agent skill with detect-agent script
and agents-seed.md (moved from engineering/setup-skills)
- implement-issue simplified: delegates agent dispatch to
tmux-launch-agent instead of inlining it
- setup-skills stripped of Section E (agent CLI config)
- New in-progress/agent-handoff skill using tmux-launch-agent
- PKM skills simplified: conversation-summary and pkm-curation
rewritten with leaner process, crit streamlined
- knowledge-gardener removed (moved to common/in-progress/)
- All READMEs updated to match
Replace the `@{prompt}` placeholder with `{prompt}` across all skill docs and
agent-cli configs, remove the deprecated `--agent` flag from the invocation
syntax, delete the now-obsolete agent-cli test file, clarify that only one
triage label should be active at a time, and defer prompt-file cleanup to the
OS/child instead of removing it immediately after tmux launch.
The skill no longer contains the contradictions this test was checking for
(prefer-reorganization phrasing, hardcoded cat requirement). Test file
served its purpose during the fix and is no longer needed.
Streamline the skill by removing tea-specific command examples. The
checks (issue exists, labels, blockers) still need to happen but
the exact commands are now at the discretion of the agent.
Also update agents-seed.md to fix arg configurations and add missing
notes about stdin piping.
- Add agents-seed.md as source of truth for supported agents
- Add Section E to setup-skills to present agent menu and validate binary on PATH
- Refactor implement-issue to read docs/agents/agent-cli.md instead of hardcoded table
- Remove detect-agent-linux.sh and detect-agent-macos.sh detection scripts
- Add mise/$SHELL -c runtime detection for agent dispatch
- Add invocation-time binary validation with clear error message
- Add test file documenting expected tmux command composition
Closes#9
- Replace `in-review` with `needs-review` in implement-issue workflow
- Document `needs-review` and `in-progress` labels in triage labels table
- Sync label definitions across setup-skills and docs/agents
- Fixed 3 broken links in root and common READMEs (pkm-curation,
forge-interaction, forge-preferences were pointing to wrong buckets)
- Moved project-context-pack from model-invoked to user-invoked
(matches its disable-model-invocation: true flag)
- Added 8 missing user-invoked and 6 missing model-invoked skills
to root and common READMEs
- Added all 9 engineering skills to engineering README
(was missing codebase-design, domain-modeling, resolving-merge-conflicts,
tdd, grill-with-docs, improve-codebase-architecture, setup-skills,
to-issues, to-prd, triage)
- Added pkm-curation to pkm README
- Updated personal README to note both skills moved elsewhere
- Added all forge skills to deprecated README
- Added .gitignore for docs/adr/
- Added docs/agents/ reference files
- Added project context cache
Break the monolithic forge-interaction skill into focused, single-responsibility
skills for better predictability and maintainability:
- Add forge-github skill for GitHub-specific interactions using `gh` CLI
- Add forge-gitea skill for Gitea-specific interactions using `tea` CLI
- Convert forge-interaction into a router that delegates to specialized skills
- Add forge-router skill with model invocation disabled for guidance only
- Move all forge-related skills to common/deprecated directory
Each specialized skill now contains its own complete decision tree, CLI usage
patterns, and authentication checks for its respective forge platform.
- Replace hardcoded open questions with runtime-derived preferences for
forge remote, CLI, and host/username detection
- Add issue label mapping reference to triage-labels.md
- Add PR title/body conventions following Conventional Commits
- Add release tagging conventions with semver format
- Add compiled wiki maintenance, retrieval lint, and packet promotion
steps to knowledge-gardener skill
- Restructure research-vault workflow with source ingest, answer filing,
and pre-completion lint pass
- Add Log.md for tracking ingest, queries, and corrections
- Enhance Claims template with provenance, supports/contradicts, and
confidence tracking
- Add Compiled Pages table and Raw Source Rules to packet template
- Enhance atomic notes with frontmatter, promotion status, and
structured links
- Distinguish source claims, synthesis, user opinions, and open
questions throughout
This implements a Karpathy-style compiled wiki approach where raw
sources stay immutable, packet notes are the maintained synthesis
layer, and reusable notes can be promoted to the main vault.
Add a new engineering skill that creates or refreshes a bounded project
memory file to serve as a navigation map for repository exploration.
The skill follows a disciplined search procedure that:
- Uses fd/rg for inventory before broad file reads
- Classifies project type with evidence citations
- Reads selectively with per-file content caps
- Indexes via tree-sitter/LSP/rg before browsing source
- Caches results in .agent/project-context.md
References the skill in README.md, common/README.md, and
common/engineering/README.md.
- Create new `pkm` directory for Personal Knowledge Management skills
- Move conversation-summary, knowledge-gardener, and research-vault from `personal` to `pkm`
- Add new crit skill for brainstorming with the CRIT framework
- Update skill descriptions for clarity and consistency
- Add audio-product-dsp, grilling, and research-engineering to model-invoked listings
- Update skill paths across README files to reflect new directory structure
- Add deprecated skills README with audio-product-dsp and research-engineering
- Add productivity skills README with grill-me, handoff, writing-great-skills, and grilling
- Move forge-preferences from user-invoked to model-invoked in personal skills
- Add research-vault skill with conversational research workflow and Obsidian packet output
- Add research packet template with Index, Sources, Synthesis, Claims, Questions, Conversation, and Glossaries
- Add personal/README.md listing personal skills
- Reorganize common/README.md paths into personal/ and productivity/ subdirectories
Add new skill definitions organized by category:
Personal:
- conversation-summary: summarize AI conversations into Obsidian notes
- knowledge-gardener: vault-aware semantic search and Zettelkasten workflows
Productivity:
- grill-me: alias for running a grilling session
- grilling: relentless interview to stress-test plans
- handoff: compact conversation into handoff document for agent continuity
- writing-great-skills: reference guide for skill authoring best practices
- glossary: domain model for skill-building terminology
Add project structure documentation (AGENTS.md) and initial README files
that organize skills by bucket and invocation type (user-invoked vs
model-invoked).
Introduce three skills:
- forge-interaction: GitHub/Gitea CLI automation for issues, PRs, releases
- forge-preferences: personal forge conventions (companion to forge-interaction)
- pkm-curation: Obsidian vault curation and note management
Include reference docs for vault conventions, agent integration, and the
invocation model that separates user-only from model-reachable skills.
gitadmin