steve 9bb5d143f7
executus CI / test (pull_request) Successful in 3m30s
fix(skillpack): address review — symlink read, git arg-injection, dup-subscribe, nil panics
Real issues from the PR review:
- security: readTree now skips symlinks (a pack with SKILL.md -> /etc/passwd
  or scripts/x -> ../../.ssh/id_rsa could read host files); covers file and
  dir symlinks, incl. within a git subpath
- security: GitSource rejects url/ref beginning with '-' (git arg injection)
  and clones with '--' separator; --filter=blob:none (blobless partial clone)
  instead of full-history clone
- correctness: Subscribe no longer swallows a non-ErrNotFound store error from
  GetByName (would create a duplicate subscription); handles *GitSource as well
  as GitSource in the URL/subpath extraction
- correctness: pinTo no longer renames a subscription, so Apply can't silently
  collide two subscriptions when an upstream pack changes its name
- validation: isKebab rejects leading/trailing/consecutive hyphens; BOM-
  prefixed SKILL.md now parses (matches the doc comment)
- robustness: Catalog/Activate/renderPackBody/Stage guard nil/malformed packs
- test cleanup: Syncer.Store field renamed Cache (collided with the Store
  interface); test NewID returns distinct ids
- tests: symlink-skip, BOM, strict-kebab, nil-pack-safety

Deferred (advisory perf, documented): PackCache stores raw trees so activation
re-parses; CheckAll is serial. Both fine at expected scale.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-04 20:41:44 -04:00

executus

⚠️ This project is vibe-coded. executus is written almost entirely by an AI coding agent (Claude), with a human steering at the design and review level rather than typing the code. That's a deliberate choice, stated up front — the same way gadfly is. Read the code before you depend on it, pin a version, and file issues if something looks off. It is offered as-is.

A batteries-included base for building LLM agent harnesses in Go. Import it, do a little wiring, and you have agentic capabilities: a bounded run loop, a tool registry with a suite of common tools, context compaction, config-driven model tiering and failover, structured output, and parallel fan-out — with sensible defaults so a brand-new project is agentic with almost no setup, and pluggable seams so a serious host can swap in its own storage, config, delivery, and tools.

executus sits strictly above majordomo — the lean LLM substrate (agent loop, canonical llm types, providers, media normalization, model parsing / failover / tiering). majordomo stays the substrate; executus is the opinionated, batteries-included layer on top. executus requires no changes to majordomo.

Status

Early. Being extracted, phase by phase, from the agent layer of mort (a Discord bot) — mort and gadfly are the first two consumers (heavy and light). See CLAUDE.md for the architecture and the extraction roadmap (P0P6).

Available today:

  • run/executus is runnable. run.Executor ties model resolution, the tool registry, majordomo's agent loop, context compaction, run-bounding, and step/audit instrumentation into one Run(ctx, RunnableAgent, inv) Result, with every host concern behind a nil-safe run.Ports (Audit/Budget/Critic/ Checkpointer/PaletteSource/Delivery/InputFiles). See examples/minimal.
  • model/ — config-driven tier resolution + failover over majordomo, with pluggable UsageSink/TraceSink and GenerateWith[T] structured output.
  • tool/ — the tool registry + 3-stage permission model + SSRF guard.
  • compact/ — the per-run context compactor.
  • lane/ — bounded worker pool with fair-share queueing (run- and provider-concurrency).
  • fanout/ — programmatic N×M swarm with bounded global + per-key concurrency.
  • config/, deliver/, identity/ — host seams (config / output / identity), each with a shipped default.
  • dispatchguard/, pendingattach/ — run-safety primitives.
  • examples/reviewer — a gadfly-shaped PR reviewer on the core only (env-config model fleet → fanout N×M swarm → model.GenerateWith[T] structured findings → consolidation), the light-tier canary; CI asserts it pulls in no battery.

Design

Two tiers in one module (go.mod = majordomo + stdlib only):

  • Core — everything a light host needs to be agentic: run loop, tool registry + common tools, model resolution, compaction, lanes, fan-out, structured output. No persistence, no scheduling.
  • Batteries (opt-in sibling packages) — persona/agent nouns, saved skills, audit, run-critic, scheduling, budgets, checkpointing. Each is nil-safe and ships a default, so you add only what you use.

Persistence that needs a real database lives in a separate nested module (contrib/store, pure-Go SQLite) so the core never drags in a DB driver — a static-binary host (gadfly) stays static.

License

TBD.

S
Description
Batteries-included base for building LLM agent harnesses in Go (above majordomo). Vibe-coded.
Readme 1.2 MiB
Languages
Go 100%