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 skills/samibs/skillfoundry/evolvenpx skills add samibs/skillfoundry --skill evolvegit clone --depth 1 https://github.com/samibs/skillfoundryWhat 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 | $0.00012 | $0.00596 |
| Opus 5 | $0.00006 | $0.00298 |
| Sonnet 5 | $0.00002 | $0.00119 |
| Haiku 4.5 | $0.00001 | $0.00060 |
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 — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/evolve - Framework Evolution from Dev-Memory
Bridge learned lessons from the dev-memory knowledge repository into claude_as framework agents.
Usage
/evolve Pull lessons and evolve agents
/evolve --dry-run Show what would change
/evolve --commit Evolve and auto-commit
/evolve --status Show evolution statistics
Instructions
You are the Evolution Engine — the bridge between accumulated project knowledge and framework intelligence.
When invoked:
Step 1: Locate dev-memory
- Check
.claude/knowledge-sync.confforDEV_MEMORY_DIR - Look in common locations (
~/dev-memory,~/projects/dev-memory) - If not found, guide user to clone it
Step 2: Run evolve.sh Execute the evolution script:
bash scripts/evolve.sh $ARGUMENTS
Step 3: Report results Show what was imported:
- Lessons learned count
- Anti-patterns imported
- Tech stack preferences
- Rules injected into
_quality-primer.md
Step 4: Propagation guidance After evolution:
To propagate to all projects:
./update.sh <project-dir>
To propagate to all registered projects:
./update.sh --scan
What evolve.sh does:
- Reads
dev-memory/global/lessons.jsonl— patterns promoted from 3+ occurrences - Reads
dev-memory/global/anti-patterns.jsonl— documented failures - Reads
dev-memory/global/tech-stack.jsonl— technology preferences - Reads
dev-memory/global/preferences.jsonl— developer preferences - Generates
agents/_learned-rules.md— full rule document for all agents - Injects top rules into
agents/_quality-primer.md"Learned Rules" section - Optionally commits to claude_as repository
The Evolution Loop:
Project A (real work)
↓ knowledge-sync.sh
dev-memory (GitHub)
↓ evolve.sh ← YOU ARE HERE
claude_as framework
↓ update.sh
All projects (improved agents)
↓ real work...
Project B learns from A's mistakes
Output Format:
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 · 100 lines · 12 tokens per session scan A 846296e155c9
evolve is a skill published in the GitHub repository samibs/skillfoundry (12 stars, last pushed 1mo ago), licensed MIT. It adds 12 tokens to every session and 596 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
ring:dispatching-workflows
Executing a phased plan in rolling waves where each phase runs as one multi-agent workflow harness: the supervisor elaborates the phase into tasks against the real landed code, launches a workflow that implements with TDD and runs mandatory in-harness review plus an adversarial contrarian pass (and researchers when…
self-improving-agent
Log learnings, errors, and corrections to .learnings/ for continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects the agent, (3) A knowledge gap is identified, (4) A better approach is found. Captures corrections, insights, errors, and feature requests; promotes broadly…
teach-me
Companion learning skill that makes a codebase legible fast. Surveys a repository, builds a cited map, runs a guided tour in dependency order, and finishes with a validated first change so the learner has actually moved the code. Use when the user runs /teach-me, or asks to learn, understand, onboard onto, get up to…
explain-diff
Explain a diff, branch, PR, patch, or review packet; use after Full-path verification when the change teaches a useful business or technical concept.
quiz-me
Companion assessment skill that grills a user on a codebase until their understanding is proven rather than assumed. Asks one question at a time, grades every answer against the real code with citations, scores 0-100 by difficulty tier, and drills the weak areas it finds. Use when the user runs /quiz-me, or asks to be…
context-manager
Use when the user asks about context organization, CLAUDE.md structure, where to store decisions or learnings, or when context feels bloated or ineffective. Teaches the 7 principles of context engineering and triages knowledge to the right destination automatically.