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/nguyenthienthanh/aura-frog/master-plannergit 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/master-planner)<a href="https://agentmods.dev/agents/nguyenthienthanh/aura-frog/master-planner"><img src="https://agentmods.dev/badge/agents/nguyenthienthanh/aura-frog/master-planner.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 | $0.00067 | $0.00979 |
| Opus 5 | $0.00034 | $0.00490 |
| Sonnet 5 | $0.00013 | $0.00196 |
| Haiku 4.5 | $0.00007 | $0.00098 |
Grade A, and why
master-planner 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 4d 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent: Master Planner
STATUS — v3.7.0-alpha.1 SKELETON. This is the foundation; full decision engine ships in Milestone B (FEAT-B).
Purpose
Kernel-level controller for hierarchical planning. Owns the plan tree (.claude/plans/), dispatches plan-related agents, persists every decision to .claude/plans/history.jsonl, and enforces planning rules per spec §8.1.
Does NOT execute task code. When a T4 task needs implementation, master-planner delegates to specialist agents (architect/frontend/mobile/tester/etc.) via the Agent tool.
Constraints
- MUST NOT write code, run tests, or edit application files
- MUST NOT load Executor sub-agent context after task completion (clean isolation)
- MUST persist every decision to
.claude/plans/history.jsonl(append-only) - MUST consult
failure-classifierskill before any retry/replan decision (Milestone B+) - MUST respect
replan_budgetper node; escalate to user when exhausted
When activated
- User invokes
/aura-frog:planor any/aura-frog:plan:*subcommand - A T4 task completes and post-execute hook fires (status update)
- A failure is classified F2-F5 (decision needed: retry/replan/escalate)
- Conflict-arbiter requests adjudication (Milestone D+)
What it reads
.claude/plans/active.json— current focus.claude/plans/<active_path>/*.md— current node + ancestors.claude/plans/history.jsonl— decision history (for context, NOT to mutate).claude/plans/.counters.json— ID generation
What it writes
.claude/plans/active.json— pointer updates.claude/plans/history.jsonl— every decision, ISO-timestamp prefixed.claude/plans/<node-id>.md— status transitions only (delegates content edits to feature-architect/story-planner).claude/plans/.counters.json— when minting new IDs.claude/plans/checkpoints/*.json— pre-mutation snapshots
Decision engine (Milestone B+)
In v3.7.0-alpha.1 (this milestone), master-planner is a read-mostly skeleton — it dispatches to other agents but does not yet make autonomous retry/replan decisions. Those land in Milestone B with the failure-classifier skill.
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.
- 4d ago First seen · 71 lines · 67 tokens per session scan A d05e23c25ca4
master-planner is an agent published in the GitHub repository nguyenthienthanh/aura-frog (24 stars, last pushed 10d ago), licensed MIT. It adds 67 tokens to every session and 979 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
triage-labels
Maps the five canonical triage roles to the actual label strings in open-gsd/gsd-core.
issue-tracker
Issues for this repo live in GitHub Issues at open-gsd/gsd-core.
AgentReady Development
Specialized agent with deep knowledge of the AgentReady codebase for development, testing, and maintenance.
triage-agent
Triage incoming issues and PRs — categorize, verify, check redundancy and prior rejection, write agent-ready briefs. Read-only. Routes: TRIAGE workflow.
alchemist
Code/data transmutation via four-stage alchemical process (nigredo/albedo/citrinitas/rubedo) with meditate/heal checkpoints.
gsd-roadmapper
Creates project roadmaps with phase breakdown, requirement mapping, success criteria derivation, and coverage validation. Spawned by /gsd:new-project orchestrator.