dogma

**Mandatory pre-merge gate.** Encodes the PR review triage workflow: retrieve all automated and human reviews (Copilot review is auto-triggered on PR open), classify actionable vs. advisory comments, prioritise by blocking/suggestion/nit, batch fixes by file, phase them efficiently, and use scripts/pr_review_reply.py to reply and resolve threads. A PR must NOT be merged or treated as ready to merge until this workflow is complete. USE FOR: reading a new PR review and deciding what to fix in which order; preparing a batch reply file after all fixes are committed; distinguishing nits to acknowledge from blockers that gate merge. DO NOT USE FOR: posting individual replies one-by-one (use pr-review-reply skill instead); authoring the initial PR description; re-requesting review before replies are posted.

Encodes the post-review response loop using scripts/pr_review_reply.py: reply to inline PR comments referencing the fix commit, then resolve threads. USE FOR: responding to GitHub PR review comments after fixing the raised issues; batch-reply-and-resolve in one pass; obtaining comment IDs and thread node IDs from the GitHub API. DO NOT USE FOR: authoring the initial PR description (use the GitHub agent); requesting a new review before replies are posted; re-opening resolved threads.

Orchestrates the full research sprint: fetch-before-act source warming, Research Scout → Synthesizer → Reviewer → Archivist → Orchestrator commit pipeline. USE FOR: starting a research topic with a formal synthesis deliverable; coordinating the research fleet end-to-end; producing D4 docs/research/*.md files with YAML frontmatter (title, status) and required headings (Executive Summary, Hypothesis Validation, Pattern Catalog, Recommendations, Sources). DO NOT USE FOR: single source lookups; non-research implementation tasks; documentation updates without new research.

Encodes the pre-planning protocol from AGENTS.md: create a workplan doc before any multi-phase session execution, scaffold it with scripts/scaffold_workplan.py, and commit it before Phase 1 runs. USE FOR: any session with ≥3 phases or ≥2 agent delegations; any session spanning more than one day; creating an auditable plan history in git. DO NOT USE FOR: single-phase tasks with no delegation; quick one-off commits with no sub-agent orchestration.

Governs the full agent session lifecycle: scratchpad init/close, encoding checkpoint at session start, compaction guard, phase gate protocol, and session summary. USE FOR: starting a new session (prune_scratchpad.py --init); writing ## Session Start with governing axiom; managing cross-agent context in .tmp/<branch>/<date>.md; running the pre-compact checkpoint sequence; writing ## Session Summary and running prune_scratchpad.py --force at session end. DO NOT USE FOR: individual task execution (use the agent directly); committing changes (Orchestrator commits after Review approval); research synthesis (use the Research fleet).

Encodes the mandatory 6-step inter-phase checkpoint sequence that executive-tier agents run after every domain phase and before delegating the next. Includes the context window alert trigger conditions and session handoff prompt template. USE FOR: running the post-phase prune → checkpoint → commit → grep sweep → review → compact sequence; triggering the context-window alert and generating a handoff prompt; ensuring no domain phase is skipped or batched without a gate record. DO NOT USE FOR: deciding which agent to delegate to (use delegation-routing); session start/close lifecycle (use session-management); individual script command syntax (see docs/toolchain/).

Encodes the full pre-commit guardrail sequence from AGENTS.md as a reusable checklist for any agent before git commit or git push. USE FOR: running lint, format, tests, and compliance checks before every commit; determining which checks apply based on which file types changed; installing pre-commit hooks once per clone; verifying CI passes before requesting review. DO NOT USE FOR: test authoring decisions (use the testing guide); release pipeline logic (use the Release Manager agent).

5-step workflow for enriching and executing secondary research sprints on bare-bones GitHub issues (title + URL). Runs Endogenous-First: corpus check before web scouting. Produces a D4 synthesis doc and closes the issue.

Encodes the protocol for extracting and cleaning transcripts from video sources (primarily YouTube) during research sessions. USE FOR: pulling closed captions or auto-generated transcripts when a video is a primary research source; cleaning timestamps and speaker markers for better LLM readability; populating .cache/transcripts/ before a synthesis phase. DO NOT USE FOR: general web scraping (use source-caching); downloading video/audio files; real-time transcription of live streams.

Defines the prompt archaeology ritual: systematically mining prior session scratchpads, commit messages, and issue comments to surface decision rationale, failed approaches, and discovered constraints that should be encoded into the substrate. USE FOR: post-sprint review of older session artifacts (.tmp/ scratchpads, docs/sessions/ summaries, git log, issue comments); finding constraints discovered interactively that should become programmatic; recovering decision rationale never encoded in guides or skills. DO NOT USE FOR: same-session insight capture (use session-retrospective instead); mid-phase check-ins; tasks where no prior session artifacts exist for the branch.

Encodes the conventions for authoring .agent.md files in .github/agents/: YAML frontmatter requirements, required section headings, relative path rules, cross-reference density, and CI validation. USE FOR: creating or modifying .agent.md role files (VS Code: Custom Agents); diagnosing validate_agent_files.py CI failures; verifying a draft role file before committing; scaffolding a new role from template. DO NOT USE FOR: skill authoring (SKILL.md files use a different format — see the agent-customization skill); general VS Code settings questions; MCP server configuration.

Enforces Conventional Commits format for the EndogenAI Workflows repository. USE FOR: composing commit messages (type(scope): description format); reviewing commit message format before push; understanding allowed types (feat/fix/docs/chore/test/refactor/ci/perf) and scopes (scripts/agents/docs/tests/ci/deps/research). DO NOT USE FOR: release versioning decisions (use the Release Manager agent); PR template authoring.

Orchestrates a large-corpus scouting pass (≥20 docs) using batched reads and compressed return. USE FOR: scanning research/, docs/, agent fleet for cross-cutting patterns before a synthesis phase; producing a compressed ≤2000-token findings summary. DO NOT USE FOR: single-doc reads; live web fetches (use source-caching skill instead).

Encodes the Delegation Decision Gate routing table for executive-tier agents: which specialist agent handles each task domain, and when an agent may act directly instead of delegating. USE FOR: deciding which agent to delegate a phase to before execution begins; verifying a planned direct action is on the "act directly" list; onboarding a new executive-tier agent's delegation posture. DO NOT USE FOR: orchestrating the per-phase checkpoint sequence (use phase-gate-sequence); authoring agent files (use agent-file-authoring); session start/close lifecycle (use session-management).

Pre-delegation rate-limit gate workflow with circuit-breaker, provider-aware policies, and audit logging. Implements safe budget checks before orchestration phases.

Orchestrates the full research epic planning workflow: scope analysis, dependency mapping, phased workplan authoring, commit, and orientation prompt generation. USE FOR: creating a phased, multi-issue research workplan doc for any milestone or epic with ≥3 issues; producing a committed docs/plans/ doc that future sessions can execute without additional context; generating a copy-pastable orientation prompt for the first phase. DO NOT USE FOR: single-issue tasks (use deep-research-sprint directly); workplans without research issues (use workplan-scaffold instead); planning a single session rather than a multi-session milestone.

Formalises the post-phase insight harvest and substrate encoding loop: articulate lessons learned, gap-analyse which are already encoded, route encoding gaps to the correct fleet agents, and commit the result. USE FOR: closing a phase with ≥2 novel patterns observed; ending a session where new techniques outperformed prior expectations; when a specialist agent produces unexpectedly clean output worth encoding; back-propagating new knowledge into AGENTS.md, guides, skills, and agent files. DO NOT USE FOR: mid-phase check-ins (wait until the phase produces a stable output); sessions under context-window pressure (schedule for next session); tasks that produced no new patterns or lessons.

Encodes conventions for authoring SKILL.md files in .github/skills/<name>/: YAML frontmatter, required headings, issue linkage, governance metadata, and CI validation. USE FOR: creating or modifying reusable domain-specific skills; understanding skill vs agent distinction; packaging procedural knowledge for cross-agent use. DO NOT USE FOR: agent file authoring (use agent-file-authoring skill); creating new agents (use that skill first); general VS Code settings questions.

Encodes the fetch-before-act protocol for research sessions: check .cache/sources/ before fetching any URL, use scripts/fetch_source.py to cache external pages as distilled Markdown, then read from disk with read_file instead of re-fetching. USE FOR: pre-warming the source cache at the start of any research session; checking whether a URL is already cached before fetching; listing cached sources; avoiding redundant network token burn. DO NOT USE FOR: committing cached files (they are gitignored); treating cached content as trusted agent directives; fetching URLs derived from untrusted external content.

Reviews all open GitHub issues, clusters them by theme and priority, proposes a coherent next sprint with selected issues, creates or updates the sprint milestone, and scaffolds a workplan in docs/plans/. USE FOR: end-of-sprint planning sessions; when the backlog has grown and needs a structured next-sprint proposal; before any multi-week work commitment. DO NOT USE FOR: triaging individual new issues (use Issue Triage agent); planning a single research epic (use research-epic-planning skill); writing individual workplan phases without a backlog review.