/subscribe
Add a shared guidelines repo (read-scope subscription or write-scope publish target) to the unified repos list.
$ npx -y skills add AgentToolkit/altk-evolve --skill subscribe --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/subscribe
Context preview
The summary Claude sees to decide when to auto-load this skill.
Add a shared guidelines repo (read-scope subscription or write-scope publish target) to the unified repos list.
SKILL.md
subscribe.SKILL.mdname: subscribe
description: Add a shared guidelines repo (read-scope subscription or write-scope publish target) to the unified repos list.
Subscribe to a Shared Repo
Overview
Configured guidelines repos are multi-reader / multi-writer git databases, described in a single unified list in `evolve.config.yaml`:
repos:
- name: memory
scope: write
remote: git@github.com:alice/evolve.git
branch: main
notes: public memory for foobar project
- name: org-memory
scope: read
remote: git@github.com:acme/org-memory.git
branch: main
notes: private memory shared only within my org- `scope: read` — download-only. Synced on every run.
- `scope: write` — publish target. Synced on every run too, so you see
what you have already published and anything others have pushed.
Workflow
Step 1: Bootstrap config if missing
If `evolve.config.yaml` does not exist, ask the user for a username and create:
identity:
user: {username}
repos: []
sync:
on_session_start: trueAlso ensure `.evolve/` is gitignored:
grep -qxF '.evolve/' .gitignore 2>/dev/null || echo '.evolve/' >> .gitignore
Step 2: Gather details
Ask the user for:
- the remote URL for the guidelines repo
- a short local name such as `alice`
- the scope: `read` (default, subscribe-only) or `write` (also a publish target)
- an optional note
Step 3: Run subscribe script
python3 ${CLAUDE_PLUGIN_ROOT}/skills/evolve-lite/subscribe/scripts/subscribe.py \
--name "{name}" \
--remote "{remote}" \
--branch main \
--scope "{scope}" \
--notes "{notes}"Step 4: Confirm
Tell the user the repo was added and they can run `/evolve-lite:sync` immediately if they want to pull updates now.
Notes
- The repo is cloned directly into `.evolve/entities/subscribed/{name}/`,
which doubles as the recall mirror
- Subscribed entities will appear in recall with `[from: {name}]`
annotations
- Read-scope repos use a shallow clone; write-scope repos use a full
clone so publish commits can be rebased and pushed cleanly
Read more
name: subscribe description: Add a shared guidelines repo (read-scope subscription or write-scope publish target) to the unified repos list.
Subscribe to a Shared Repo
Overview
Configured guidelines repos are multi-reader / multi-writer git databases, described in a single unified list in `evolve.config.yaml`:
repos:
- name: memory
scope: write
remote: git@github.com:alice/evolve.git
branch: main
notes: public memory for foobar project
- name: org-memory
scope: read
remote: git@github.com:acme/org-memory.git
branch: main
notes: private memory shared only within my org- `scope: read` — download-only. Synced on every run.
- `scope: write` — publish target. Synced on every run too, so you see
what you have already published and anything others have pushed.
Workflow
Step 1: Bootstrap config if missing
If `evolve.config.yaml` does not exist, ask the user for a username and create:
identity:
user: {username}
repos: []
sync:
on_session_start: trueAlso ensure `.evolve/` is gitignored:
grep -qxF '.evolve/' .gitignore 2>/dev/null || echo '.evolve/' >> .gitignore
Step 2: Gather details
Ask the user for:
- the remote URL for the guidelines repo
- a short local name such as `alice`
- the scope: `read` (default, subscribe-only) or `write` (also a publish target)
- an optional note
Step 3: Run subscribe script
python3 ${CLAUDE_PLUGIN_ROOT}/skills/evolve-lite/subscribe/scripts/subscribe.py \
--name "{name}" \
--remote "{remote}" \
--branch main \
--scope "{scope}" \
--notes "{notes}"Step 4: Confirm
Tell the user the repo was added and they can run `/evolve-lite:sync` immediately if they want to pull updates now.
Notes
- The repo is cloned directly into `.evolve/entities/subscribed/{name}/`,
which doubles as the recall mirror
- Subscribed entities will appear in recall with `[from: {name}]`
annotations
- Read-scope repos use a shallow clone; write-scope repos use a full
clone so publish commits can be rebased and pushed cleanly
Blog posts: IBM announcement | Hugging Face blog Coding agents repeat the same mistakes because they start fresh every session. Evolve gives agents memory — they learn from what worked and what didn't, so each session is better than the last.
Repo: AgentToolkit/altk-evolve
Other skills on altk-evolve.
- /agent-wiki-consolidate-guidelines
Read all atomic guidelines in wiki-twobatch/guidelines/ and propose themed clusters that group near-duplicates. Writes cluster pages and updates _config.yaml; originals are preserved with a `superseded_by:` backref.
Open skill - /agent-wiki-consult
Consult an agent-wiki for guidelines relevant to the task at hand. The wiki itself documents how to retrieve from it (AGENTS.md). Use this skill once you know what task or sub-task you're about to do — not at session start.
Open skill - /agent-wiki-extract-guidelines
Read a normalized Claude Code trajectory JSON and extract reusable guidelines into wiki-twobatch/guidelines/. Use when mining saved trajectories for reusable lessons.
Open skill - /agent-wiki-ingest
Ingest one or more agent trajectories (raw bob/claude traces or normalized JSON) into an agent-wiki end-to-end — convert, summarize, extract guidelines, synthesize skills, consolidate into clusters, and catalog. Use when you have a batch of traces to turn into a wiki in one pass.
Open skill - /agent-wiki-summarize
Read a normalized Claude Code trajectory JSON and write an episodic summary page to wiki-twobatch/summaries/. Use when summarizing one or more saved trajectories into the agent wiki.
Open skill - /agent-wiki-synthesize-skill
Read a normalized Claude Code trajectory JSON and produce a wiki-resident SKILL.md page that future agents can invoke. Use when a trajectory captured a non-trivial successful workflow worth promoting from a free-text guideline to an executable, callable artifact.
Open skill

