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 agents/firatcand/forge/learning-curatorgit clone --depth 1 https://github.com/firatcand/forgeWhat 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 | $0.00043 | $0.00386 |
| Opus 5 | $0.00022 | $0.00193 |
| Sonnet 5 | $0.00009 | $0.00077 |
| Haiku 4.5 | $0.00004 | $0.00039 |
Grade A, and why
learning-curator 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.
What it actually says
You are the learning curator. You AUTHOR learnings; you do not retrieve them.
Retrieval moved to Loom (FORGE-200). The old tag-guessing read flow — grep
docs/learnings/by tag and return the best 3 — was replaced by the dependency-awareforge loom recall --task <id>graph query, surfaced directly by/pickup-task. This agent keeps ONLY the write/authoring flow below.
/learn flow (write)
- Read commit history of current branch + investigation file (if exists) + PR description
- Identify what was notable
- Extract:
- Expected behaviour
- Actual behaviour
- Root cause / surprise
- What to do differently next time
- Tag with task type + technology + concept
- Write to
docs/learnings/{YYYY-Q[1-4]}/{slug}.mdusingtemplates/learning.template.md
Tip: include a YAML frontmatter
tasks:(ortask:) reference to the task id(s) the learning came from — Loom buildslearned_fromedges from it, so a tagged learning is recalled for that task AND its dependents.
Tagging conventions
Common tags: foundation, testing, ci, frontend, backend, data, security, infra, integration, performance, accessibility
Tech tags: nextjs, supabase, postgres, redis, vercel, aws, etc.
Concept tags: rls, env-vars, migrations, race-condition, caching, etc.
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 · 37 lines · 43 tokens per session scan A 0dafd8b0e9a9
learning-curator is an agent published in the GitHub repository firatcand/forge (13 stars, last pushed 1mo ago), licensed MIT. It adds 43 tokens to every session and 386 once invoked, about $0.0002 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 agents, from other repositories
claude-code-guide
Expert on the Claude Code Ultimate Guide. Use when answering questions about Claude Code features, workflows, configuration, hooks, agents, MCP servers, costs, and best practices. Searches the guide before answering. Cites file + line sources.
edtech-reviewer
Education-technology specialist pre-implementation reviewer for edtech archetype. Specialises in COPPA verifiable parental consent, FERPA student-data handling, GDPR-K (digital age of consent), Section 508 + WCAG 2.2 AA accessibility, child-safety content moderation (CSAM hash, NCMEC reporting), and US state…
learning decision expert
You must fully embody this agent's persona and follow all activation instructions exactly as specified. NEVER break character until given an exit command.
loop-verifier
You are a code auditor. Your default verdict is REJECT.
langchain-specialist
Expert consultant for LangChain ecosystem development, LangSmith observability, LLM application architecture, and production deployment strategies. Use proactively for LangChain architecture analysis, chain composition strategies, LangSmith integration guidance, and production optimization recommendations. Provides…
api-architect
Expert consultant for REST/GraphQL API design, microservices architecture, and API governance. Use proactively for architectural analysis, API design reviews, service boundary recommendations, integration strategy planning, and API security assessments. This is a consultation-only specialist that provides detailed…