EndogenAI

EndogenAI/dogma

Hook Claude Code

Runs when a session starts, executing prune_scratchpad.py via uv with --init. From EndogenAI/dogma.

EndogenAI/dogma

Hook Claude Code

Runs before the context is compacted, executing prune_scratchpad.py via uv with --check-size. From EndogenAI/dogma.

EndogenAI/dogma

Settings file Claude Code

Agent settings declaring 5 hook events (SessionStart, Stop, PreCompact, PostCompact) and 9 allowed tools.

EndogenAI/dogma

Agent

This file is a subdirectory-level governance override. All general constraints are now centralized in the root AGENTS.md.

EndogenAI/dogma

Agent

Evaluate context governance options — AFS context layers, semantic memory isolation, and scratchpad vs repo-memory tradeoffs — and design context layering conventions.

EndogenAI/dogma

Agent

Guide companion-repo onboarding via adoptwizard.py and the cookiecutter template; produce a configured client-values.yml and AGENTS.md for new adopters.

EndogenAI/dogma

Agent

Audit uv.lock and pyproject.toml for known CVEs, outdated packages, and version conflicts; output a structured compatibility report.

EndogenAI/dogma

Agent

Synthesizes customer insights, tracks consulting pipeline, informs pricing strategy, and prepares deal mechanics. Operates with discovery/facilitation scope in Month 1; decision authority expands based on Conor's confidence. All recommendations require Conor approval during Phase 1 (months 1–3).

EndogenAI/dogma

Agent

Watch CI run history for recurring failure patterns, triage flaky tests, identify slow steps, and surface actionable fix suggestions for the GitHub Actions pipeline.

EndogenAI/dogma

Agent

Defines messaging framework, proposes content calendar, establishes brand voice, and manages community messaging. Operates with facilitation scope in Month 1; autonomy over messaging expands in Month 2–3 based on Conor + Sheela validation. All public-facing content requires approval in Phase 1.

EndogenAI/dogma

Agent

Aggregate GitHub community health signals — Stars, issue response time, PR velocity, contributor retention — and produce periodic health reports for project maintainers.

EndogenAI/dogma

Agent

Validate proposed changes against MANIFESTO.md axioms — flag over-interactivity, unnecessary token burn, and Programmatic-First violations.

EndogenAI/dogma

Agent

Manage docs/research/OPENRESEARCH.md as a living queue — track status, retire completed items, and propose new seed questions from synthesis gaps.

EndogenAI/dogma

Agent

Execute a recursive, hypothesis-driven deep dive research workflow — orchestrating corpus scanning, web scouting, source fetching, sprint synthesis, bibliography enrichment, and academic paper generation. Instantiates the workflow defined in docs/guides/deep-research.md.

EndogenAI/dogma

Agent

Plan developer relations content — blog cadence, tutorial pipeline, Discussions announcements, and DevEx narrative for an agent-first open-source project.

EndogenAI/dogma

Agent

Audit docs/research/ for D4 heading gaps, dead source stubs, missing frontmatter, and validatesynthesis compliance. Advisory-only; never edits docs directly.

EndogenAI/dogma

Agent

Audit Python environment consistency across .python-version, uv.lock, pyproject.toml, and CI matrix — flag compatibility drift and suggest remediation.

EndogenAI/dogma

Agent

Design and implement non-agent automation — file watchers, pre-commit hooks, CI tasks, and VS Code task definitions. First escalation point for anything that should run without an agent.

EndogenAI/dogma

Agent

Maintain and evolve all project documentation — encoding dogmatic values, guiding axioms, and methodology across every documentation layer.

EndogenAI/dogma

Agent

Manage the agent fleet — create, audit, update, and deprecate .agent.md files and fleet documentation to maintain standards compliance.

EndogenAI/dogma

Agent

Coordinate multi-workflow sessions spanning research, docs, scripting, and fleet changes — sequence executive agents and maintain session coherence.

EndogenAI/dogma

Agent

Decompose complex multi-step requests into structured plans with phases, gates, agent assignments, and dependency ordering before any execution begins.

EndogenAI/dogma

Agent

Maintain the health of the repository as an open-source resource — issues, labels, milestones, changelog, contributing docs, and community standards.

EndogenAI/dogma

Agent

Orchestrate research sessions end-to-end — delegate to the research fleet, synthesize outputs, and spawn new area-specific research agents as needed.