Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/hknc/claude-evolve/evolvegit clone --depth 1 https://github.com/hknc/claude-evolveWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00070 | $0.02140 |
| Opus 5 | $0.00035 | $0.01070 |
| Sonnet 5 | $0.00014 | $0.00428 |
| Haiku 4.5 | $0.00007 | $0.00214 |
Grade A, and why
evolve scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 254 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/evolve Command
Architecture Note
CRITICAL: Commands handle ALL user interaction. Subagents CANNOT use AskUserQuestion.
Subagents invoked via Task() are stateless and cannot pause for user input. The command must collect all user choices BEFORE spawning the agent.
Routing
| Command | Handler |
|---|---|
/evolve (bare) |
Task(claude-evolve:evolve-toolkit-manager) with action="status" |
/evolve init |
See Init Flow below |
/evolve help |
Read file ${CLAUDE_PLUGIN_ROOT}/commands/help.md and output its content EXACTLY. Do NOT search for or scan any directories. |
/evolve status |
Task(claude-evolve:evolve-toolkit-manager) with action="status" |
/evolve release |
See Release Flow below |
/evolve migrate |
Task(claude-evolve:evolve-toolkit-manager) with action="migrate" |
/evolve switch {name} |
Task(claude-evolve:evolve-toolkit-manager) with action="switch", name="{name}" |
/evolve context |
Task(claude-evolve:evolve-context-detector) |
/evolve learn |
Same as /learn command — execute the two-phase interactive flow from ${CLAUDE_PLUGIN_ROOT}/commands/learn.md |
/evolve audit |
Task(claude-evolve:evolve-setup-auditor) |
/evolve consolidate |
Task(claude-evolve:evolve-setup-auditor) with action="consolidate" |
/evolve prepare |
Use prepare-task skill |
/evolve understanding |
List all understanding files (see below) |
Understanding Command (/evolve understanding)
Show the user's learned patterns and learning activity. Run these bash commands and display results:
TOOLKIT_NAME=$(cat $HOME/.claude-evolve/active 2>/dev/null)
TOOLKIT_PATH=$HOME/.claude-evolve/toolkits/$TOOLKIT_NAME
UNDERSTANDING_PATH=$TOOLKIT_PATH/understanding
Output format:
## Your Understanding
**Universal patterns:** {UNDERSTANDING_PATH}/understanding.md
{Show first 15 lines of content or "Empty - run /learn then /reflect to build"}
**Project-specific:**
{List all files in {UNDERSTANDING_PATH}/projects/*.md with their first line}
{Or "None yet - run /reflect after /learn sessions to build project patterns"}
**Learning history:**
{Count events in $TOOLKIT_PATH/history/events.json or "No learnings yet"}
{Show last 5 events (summary + timestamp) if events exist}
**Toolkit components:**
{Count agents in $TOOLKIT_PATH/agents/*.md}
{Count skills in $TOOLKIT_PATH/skills/*/SKILL.md}
{Count rules in $TOOLKIT_PATH/rules/*.md}
**Pending signals:** {count from $HOME/.claude-evolve/signals/*.json}
{If count > 0: "Run /learn to capture pending signals"}
{If count = 0: "No pending signals"}
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 254 lines · 70 tokens per session scan A eeed753ba42c
evolve is a command published in the GitHub repository hknc/claude-evolve (8 stars, last pushed 7mo ago), licensed MIT. It adds 70 tokens to every session and 2,140 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.