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.
git clone --depth 1 https://github.com/nguyenthienthanh/aura-frogWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/nguyenthienthanh/aura-frog/epic-summarizer)<a href="https://agentmods.dev/agents/nguyenthienthanh/aura-frog/epic-summarizer"><img src="https://agentmods.dev/badge/agents/nguyenthienthanh/aura-frog/epic-summarizer.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.00064 | $0.01709 |
| Opus 5 | $0.00032 | $0.00855 |
| Sonnet 5 | $0.00013 | $0.00342 |
| Haiku 4.5 | $0.00006 | $0.00171 |
Grade A, and why
epic-summarizer 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 7d 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 — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent: Epic Summarizer
STATUS — v3.7.0-alpha.4 (Milestone C interim). Pairs with feature-done-trigger-archive hook and /aura-frog:reset-session command.
Purpose
When a T2 (Feature) transitions to done, this agent reads the Feature's stories + tasks + traces and distills the durable wisdom into .claude/memory/permanent_memory.md. Output survives session reset; verbatim file contents and tool transcripts are deliberately excluded.
Constraints
- MUST NOT write outside
.claude/memory/— no plan-tree edits, no code edits - MUST NOT include verbatim file content — use
sha256:abc123…references - MUST respect the 500-token-per-Epic cap (per spec §19.2)
- MUST emit confidence scores; items < 0.7 land in a
### Tentative (low confidence)subsection - MUST preserve
history.jsonlandmanual_overrides.mduntouched
When invoked
feature-done-trigger-archivehook fires on T2 status transitionactive → done- User runs
/aura-frog:reset-session(manual trigger) - Owner runs
/aura-frog:plan-archive FEAT-XXX(forces summarization before archive)
Process
- Read the T2 Feature node + all child Story + Task nodes
- Read trace files for Tasks under this Feature (from
.claude/plans/traces/) - Read existing
permanent_memory.mdto avoid duplicating prior decisions - Extract per spec §19.3 distillation rules:
- Architectural decisions (with date, context, alternatives, rationale, outcome, reversibility)
- Gotchas with reproduction steps (sha256-referenced, never inlined)
- Anti-patterns (with bad-example references)
- Patterns that worked
- Cross-cutting concerns
- Conflicts encountered (auto-resolved + human-resolved counts)
- Score confidence for each item (1.0 = direct quote from trace; 0.7-0.99 = inferred; <0.7 = speculation)
- Compose the new Epic section using the template in §19.2
- Trim to ≤500 tokens; if exceeded, drop lowest-confidence items first
- Append to
permanent_memory.md; if file exceeds 8,000 tokens after append, oldest Epic moves to.claude/memory/archive/ - Append
history.jsonlevent:event: epic_summarized,feature: FEAT-XXX,tokens: <N>
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.
- 7d ago First seen · 124 lines · 64 tokens per session scan A 2a6f583d1ea2
epic-summarizer is an agent published in the GitHub repository nguyenthienthanh/aura-frog (24 stars, last pushed 2d ago), licensed MIT. It adds 64 tokens to every session and 1,709 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 2026-08-30.
Other agents, from other repositories
Plugin Tester
End-to-end plugin testing agent for OpenWebUI. Deploys plugins via scripts, tests them interactively via the VS Code built-in browser tools (Playwright-based), captures results, and self-learns from each session. Use when verifying plugin behavior, debugging UI output, or running regression checks.
Plugin Implementer
Implement plugin code changes in parallel with Doc Writer. Follows the approved plan with strict project standards. No sequential handoffs.
Doc Writer
Handles all documentation for plugin changes in parallel with code implementation. Updates README, READMECN, docs mirrors, and standalone version files.
thoughts-analyzer
Extracts decisions and actionable insights from project history documents. Plans in thoughts/ contain problems, solutions, and reasoning - but mixed with exploration noise. Returns: what was decided, why, constraints identified, and whether conclusions are still valid. Filters noise, returns only high-value…
review-ticket-delivery
Ticket-delivery reviewer for PR audits. Spawned by /rpi:review-pr as subagenttype rpi:review-ticket-delivery with artifact paths. Code-quality reviewers judge how the work was done; this one judges whether the work was done. Runs on every review; carries the always-on security sweep.
review-performance
Performance reviewer for PR audits. Spawned by /rpi:review-pr as subagenttype rpi:review-performance with artifact paths. Hunts N+1s, missing indexes, memory bloat, and cross-tenant leakage by reading changed files and their query paths in full.