Local Skill Iteration¶

Working on a skill or pattern locally — without rebuilding the agent-smith binary, without releasing a new version of agentsmith-skills.

Quick path: clone the catalog repo, point at it¶

# Clone the catalog repo somewhere outside the agent-smith working tree
git clone https://github.com/holgerleichsenring/agent-smith-skills /path/to/agent-smith-skills

# Point your local config at it
# agentsmith.yml:
skills:
  source: path
  path: /path/to/agent-smith-skills

Edits to /path/to/agent-smith-skills/skills/**/SKILL.md are picked up by the next pipeline run — no rebuild, no restart needed (the loader reads SKILL.md on each load).

Variant: bind-mount in docker-compose¶

# deploy/docker-compose.yml additions
services:
  agent-smith:
    volumes:
      - ../agent-smith-skills:/var/lib/agentsmith/skills:ro
    environment:
      AGENTSMITH_SKILLS_DIR: /var/lib/agentsmith/skills

Together with skills.source: path in agentsmith.yml, this lets you edit SKILL.md on your host and have the container pick up the change immediately.

Test runs¶

The test suite uses ./test-skills/ populated by scripts/fetch-skills.sh. For local-iteration tests, point at your working copy instead:

export AGENTSMITH_TEST_SKILLS_DIR=/path/to/agent-smith-skills
dotnet test

Releasing changes¶

Once the local iteration looks good:

Push your changes to the agentsmith-skills fork or branch.
Tag vX.Y.Z. The release.yml workflow builds the deterministic tarball and publishes the release.
Bump skills.version in your agent-smith production config to the new tag.
Roll the server. The new tag mismatches the cached .pulled marker and triggers a fresh pull.

Edge channel (tag: edge) tracks main and is rebuilt on every push — useful for smoke-testing in a staging environment without minting a release.